ic2s2 visualisation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Take the accepted talks from ic2s2 and create a co-authorship graph, and a person - concept graph in neo4j. \n",
    "\n",
    "The accepted talks are available at https://ic2s2.org/2017/elements/accepted_talks.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from neo4jrestclient.client import GraphDatabase\n",
    "from bs4 import BeautifulSoup\n",
    "from collections import defaultdict"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Let's get the contents of that webpage and stash it locally.\n",
    "\n",
    "```\n",
    "$ curl -0 https://ic2s2.org/2017/elements/accepted_talks.html > accepted_talks.html\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/ianm/anaconda/lib/python3.6/site-packages/bs4/__init__.py:181: UserWarning: No parser was explicitly specified, so I'm using the best available HTML parser for this system (\"lxml\"). This usually isn't a problem, but if you run this code on another system, or in a different virtual environment, it may use a different parser and behave differently.\n",
      "\n",
      "The code that caused this warning is on line 193 of the file /Users/ianm/anaconda/lib/python3.6/runpy.py. To get rid of this warning, change code that looks like this:\n",
      "\n",
      " BeautifulSoup([your markup])\n",
      "\n",
      "to this:\n",
      "\n",
      " BeautifulSoup([your markup], \"lxml\")\n",
      "\n",
      "  markup_type=markup_type))\n"
     ]
    }
   ],
   "source": [
    "# check we can access the file and process it with Beautiful soup \n",
    "filename = \"accepted_talks.html\"\n",
    "html = open(filename,\"r\").read()\n",
    "soup = BeautifulSoup(html)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Goodness, the HTML is not very well structured, we are going to have to do some inference here around what different types of content are! \n",
    "\n",
    "All of the titles are in bold. All of the authors are siblings of content in bold, let's have a think! "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# we want to extract the authors and titles of each talk \n",
    "talks = []\n",
    "section = soup.find('section', {'class': 'section', 'id': 'colgne'})\n",
    "div = section.find(\"div\")\n",
    "btags = div.findAll(\"b\")\n",
    "for btag in btags:\n",
    "    title = btag.contents\n",
    "    authors = btag.next_sibling.next_sibling\n",
    "    author_list = authors.replace(\" and \", \",\").split(\",\") \n",
    "    talks.append((title, author_list))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# now we have extracted the titles and authors from the page, I noticed that the talks are listed twice on the page, \n",
    "# lets get a unique set of talks\n",
    "unique_talks = []\n",
    "for talk in talks:\n",
    "    if talk not in unique_talks:\n",
    "        unique_talks.append(talk)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# get a dumb set of stopwords\n",
    "stopwords = open(\"SmartStoplist.txt\").readlines()\n",
    "stopwords_stripped = list(map(lambda x: x.strip(), stopwords))\n",
    "strip_characters = [\":\", \"?\", \"-\", \"'\", \"`\", \";\", \",\", \".\", \"\\\"\" ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def clean_word(word):\n",
    "    \"\"\"\n",
    "    Ok, we are going to do a bit of claning up here, just a quick job, namely\n",
    "    normalise words to lowercase \n",
    "    get rid of some punctuation \n",
    "    if the word is not on our list, strip s from the end to get rid of plural variants (so dumb, but works right now)\n",
    "    \"\"\"\n",
    "    candidate_word = word.lower().strip()\n",
    "    current_word = candidate_word\n",
    "    for char in strip_characters:\n",
    "        current_word = current_word.replace(char, '')\n",
    "    if current_word not in [\"bias\", \"embeddedne\", \"ubiquitous\", \"analysis\", \"process\", \"strategie\", \"properties\"]:\n",
    "        cleaned_word = current_word.rstrip(\"s\") # dumb way of getting rid of plurals, OK for a short analysis \n",
    "        return cleaned_word\n",
    "    else:\n",
    "        return current_word"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_keywords(title):\n",
    "    \"\"\"\n",
    "    go through all of the words in a titls, and drop words in our small stop word list\n",
    "    that way we can try to generate words that might look at bit like keywords for the talks. \n",
    "    \"\"\"\n",
    "    keywords = []\n",
    "    for word in title.split():\n",
    "        cleaned_word = clean_word(word)\n",
    "        if cleaned_word not in stopwords_stripped:\n",
    "            keywords.append(cleaned_word)\n",
    "    return keywords"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "School segregation in the digital space ['school', 'segregation', 'digital', 'space']\n",
      "Bounded confidence in extreme opinion evolution : an online experiment ['bounded', 'confidence', 'extreme', 'opinion', 'evolution', '', 'online', 'experiment']\n",
      "Do Twitter Data reflect the Link between economic and cultural Modernization? ['twitter', 'data', 'reflect', 'link', 'economic', 'cultural', 'modernization']\n",
      "Modeling Echo Chambers on Social Media ['modeling', 'echo', 'chamber', 'social', 'media']\n",
      "New Techniques for Privacy-preserving Record Linkage of Large-scale Social Science Data Sets ['technique', 'privacypreserving', 'record', 'linkage', 'largescale', 'social', 'science', 'data', 'set']\n",
      "Ideological Creation of Market in Socialist China: Change in Economic Rhetoric in the People’s Daily, 1946-2003 ['ideological', 'creation', 'market', 'socialist', 'china', 'change', 'economic', 'rhetoric', 'people’', 'daily', '19462003']\n",
      "The Network Dynamics of Social Influence in the Wisdom of Crowds ['network', 'dynamic', 'social', 'influence', 'wisdom', 'crowd']\n",
      "Why We Read Wikipedia ['read', 'wikipedia']\n",
      "Can Network Theory-based Targeting Increase Technology Adoption? ['network', 'theorybased', 'targeting', 'increase', 'technology', 'adoption']\n",
      "Moral Foundations of Political Discourse: Comparative Analysis of the Speech Records of the US Congress and the Japanese Diet ['moral', 'foundation', 'political', 'discourse', 'comparative', 'analysis', 'speech', 'record', 'congre', 'japanese', 'diet']\n",
      "The Effect of Missing Translations on Analyses of European Legislative Speech ['effect', 'missing', 'translation', 'analyse', 'european', 'legislative', 'speech']\n",
      "Sampling from Social Networks with Attributes ['sampling', 'social', 'network', 'attribute']\n",
      "The fall of the empire: The Americanization of English ['fall', 'empire', 'americanization', 'english']\n",
      "An Analysis of Sponsored Content on Twitter ['analysis', 'sponsored', 'content', 'twitter']\n",
      "Psychological and Personality Profiles of Political Extremists ['psychological', 'personality', 'profile', 'political', 'extremist']\n",
      "Who Cares About the Overhead? Measuring the Efficiency in the Charity Marketplace ['care', 'overhead', 'measuring', 'efficiency', 'charity', 'marketplace']\n",
      "The Spread of Fake News on Twitter by Voters During the 2016 U.S. General Election  ['spread', 'fake', 'twitter', 'voter', '2016', 'general', 'election']\n",
      "Statistical Scaling Meets Classic Content Analysis: Using Bayesian IRT-Modelling to Infer Latent Variables from Hand-Coded Texts ['statistical', 'scaling', 'meet', 'classic', 'content', 'analysis', 'bayesian', 'irtmodelling', 'infer', 'latent', 'variable', 'handcoded', 'text']\n",
      "The Memory Remains:\\\\ Understanding Collective Memory in the Digital Age ['memory', 'remains\\\\\\\\', 'understanding', 'collective', 'memory', 'digital', 'age']\n",
      "The Wisdom of Dynamic Networks: Selective Social Learning and the Adaptive Wisdom of Crowds ['wisdom', 'dynamic', 'network', 'selective', 'social', 'learning', 'adaptive', 'wisdom', 'crowd']\n",
      "Perceiving and Acting: How Preferences and Perceptions of Cultural Norms Relate to Normative Compliance ['perceiving', 'acting', 'preference', 'perception', 'cultural', 'norm', 'relate', 'normative', 'compliance']\n",
      "Using named entities for computer-automated verbal deception detection ['named', 'entitie', 'computerautomated', 'verbal', 'deception', 'detection']\n",
      "Inducing Egocentric Networks with Privacy Settings: Evidence from Integrated Social Media and Representative Survey Data ['inducing', 'egocentric', 'network', 'privacy', 'setting', 'evidence', 'integrated', 'social', 'media', 'representative', 'survey', 'data']\n",
      "Exploring Emotions in Online Movie Reviews ['exploring', 'emotion', 'online', 'movie', 'review']\n",
      "The Hostility Bonus: How Negativity Dominates American Political Communication ['hostility', 'bonu', 'negativity', 'dominate', 'american', 'political', 'communication']\n",
      "Understanding social influence and inactivity patterns in the Twitter society ['understanding', 'social', 'influence', 'inactivity', 'pattern', 'twitter', 'society']\n",
      "Promoting Educational Attainment at Scale with Psychological Interventions ['promoting', 'educational', 'attainment', 'scale', 'psychological', 'intervention']\n",
      "Methods for Acquiring, Analyzing, and Visualizing Complex Street Networks ['method', 'acquiring', 'analyzing', 'visualizing', 'complex', 'street', 'network']\n",
      "The Polarized Network of Skill Complementarity ['polarized', 'network', 'skill', 'complementarity']\n",
      "Transparency and Re-Use Patterns of Open Digital Crowd Contests: Linking Empirical and Computational Approaches ['transparency', 'reuse', 'pattern', 'open', 'digital', 'crowd', 'contest', 'linking', 'empirical', 'computational', 'approache']\n",
      "Identifying Sink and Conduit Offshore Financial Centers in the Global Corporate Ownership Network ['identifying', 'sink', 'conduit', 'offshore', 'financial', 'center', 'global', 'corporate', 'ownership', 'network']\n",
      "Insurance Fraud Detection: A Network Modeling and Behavioral Analytics Approach ['insurance', 'fraud', 'detection', 'network', 'modeling', 'behavioral', 'analytic', 'approach']\n",
      "Political Ideology and Affective Attitudes Towards Mobility Innovation: Comparing Survey Responses and Twitter Data ['political', 'ideology', 'affective', 'attitude', 'mobility', 'innovation', 'comparing', 'survey', 'response', 'twitter', 'data']\n",
      "Visibility of minorities in social networks ['visibility', 'minoritie', 'social', 'network']\n",
      "Estimating the possibility of trajectory-based reidentification for fusion of large-scale urban mobility datasets ['estimating', 'possibility', 'trajectorybased', 'reidentification', 'fusion', 'largescale', 'urban', 'mobility', 'dataset']\n",
      "Sharing Strategies: Optimal networks for team collaboration and problem solving ['sharing', 'strategie', 'optimal', 'network', 'team', 'collaboration', 'problem', 'solving']\n",
      "Statistical properties of fluctuations of time series representing appearances of words in nationwide blog data and their applications ['statistical', 'propertie', 'fluctuation', 'time', 'serie', 'representing', 'appearance', 'word', 'nationwide', 'blog', 'data', 'application']\n",
      "Gender Outcome Disparities in an Online Community of Designers ['gender', 'outcome', 'disparitie', 'online', 'community', 'designer']\n",
      "Modelling discourse patterns in small groups of competing and co-operating members ['modelling', 'discourse', 'pattern', 'small', 'group', 'competing', 'cooperating', 'member']\n",
      "Social Bots in Japan’s 2014 General Election: Preliminary Results from a Corpus-Linguistic and Qualitative Study of Computational Propaganda on Twitter ['social', 'bot', 'japan’', '2014', 'general', 'election', 'preliminary', 'result', 'corpuslinguistic', 'qualitative', 'study', 'computational', 'propaganda', 'twitter']\n",
      "Large fluctuation in social tagging ecosystems ['large', 'fluctuation', 'social', 'tagging', 'ecosystem']\n",
      "Using Agent Based Modeling to Interpret Underlying Factors of Underrepresentation of Minorities in Hollywood Films ['agent', 'based', 'modeling', 'interpret', 'underlying', 'factor', 'underrepresentation', 'minoritie', 'hollywood', 'film']\n",
      "From Migration Corridors to Clusters: The Value of Google+ Data for Migration Studies ['migration', 'corridor', 'cluster', 'google+', 'data', 'migration', 'studie']\n",
      "Linguistic neighbourhoods: Explaining cultural borders on Wikipedia through multilingual co-editing activity ['linguistic', 'neighbourhood', 'explaining', 'cultural', 'border', 'wikipedia', 'multilingual', 'coediting', 'activity']\n",
      "Dynamic Networks of News Consumption: Online Traffic during the 2016 Brexit Campaign in the UK ['dynamic', 'network', 'consumption', 'online', 'traffic', '2016', 'brexit', 'campaign', 'uk']\n",
      "Evolving affiliations amongst UK politicians ['evolving', 'affiliation', 'uk', 'politician']\n",
      "Quantifying patterns of research interest evolution ['quantifying', 'pattern', 'research', 'interest', 'evolution']\n",
      "LabVanced: A Unified JavaScript Framework for Online Studies ['labvanced', 'unified', 'javascript', 'framework', 'online', 'studie']\n",
      "Moral Foundations in Online Conversations ['moral', 'foundation', 'online', 'conversation']\n",
      "A Topic-Sentiment Evolution Model to Analyze Chinese National Image from Online News ['topicsentiment', 'evolution', 'model', 'analyze', 'chinese', 'national', 'image', 'online']\n",
      "Inter-modular connectivity creates structural instability in complex contagion ['intermodular', 'connectivity', 'create', 'structural', 'instability', 'complex', 'contagion']\n",
      "Modeling collective attention on promoted content in Wikipedia ['modeling', 'collective', 'attention', 'promoted', 'content', 'wikipedia']\n",
      "Detecting Political Bots on Russian Twitter ['detecting', 'political', 'bot', 'russian', 'twitter']\n",
      "Celebrity Suicides and the Werther Effect. An Analysis of 34 OECD Countries using Wikipedia’s Celebrity Suicide Data ['celebrity', 'suicide', 'werther', 'effect', 'analysis', '34', 'oecd', 'countrie', 'wikipedia’', 'celebrity', 'suicide', 'data']\n",
      "Effects of temporal correlations in social multiplex networks ['effect', 'temporal', 'correlation', 'social', 'multiplex', 'network']\n",
      "What Difference Does Difference Make? A Language-Based Model of Cultural Heterogeneity and Firm Performance ['difference', 'doe', 'difference', 'make', 'languagebased', 'model', 'cultural', 'heterogeneity', 'firm', 'performance']\n",
      "Market uncertainty and the network structure of transaction sequences in the EU interbank money market ['market', 'uncertainty', 'network', 'structure', 'transaction', 'sequence', 'eu', 'interbank', 'money', 'market']\n",
      "Platform effects on social media: Topic salience in a representative survey, Facebook and Twitter during an election campaign ['platform', 'effect', 'social', 'media', 'topic', 'salience', 'representative', 'survey', 'facebook', 'twitter', 'election', 'campaign']\n",
      "When to target hubs? Strategic Diffusion in Complex Networks ['target', 'hub', 'strategic', 'diffusion', 'complex', 'network']\n",
      "Understanding Regional Economy by Decomposing Urban Labor Systems ['understanding', 'regional', 'economy', 'decomposing', 'urban', 'labor', 'system']\n",
      "Order without law: Reputation promotes cooperation in a cryptomarket for illegal drugs ['order', 'law', 'reputation', 'promote', 'cooperation', 'cryptomarket', 'illegal', 'drug']\n",
      "How North Korea Views China: Applying Sentiment Analysis to Korean Central News Agency Reports ['north', 'korea', 'view', 'china', 'applying', 'sentiment', 'analysis', 'korean', 'central', 'agency', 'report']\n",
      "Trust Networks: Topology, Dynamics and Measurements ['trust', 'network', 'topology', 'dynamic', 'measurement']\n",
      "Face-to-BMI: Using Computer Vision to Study Body Weight in Social Media Profile Pictures ['facetobmi', 'computer', 'vision', 'study', 'body', 'weight', 'social', 'media', 'profile', 'picture']\n",
      "Mapping evolving uncertainty in a complex engineering design project with email mining ['mapping', 'evolving', 'uncertainty', 'complex', 'engineering', 'design', 'project', 'email', 'mining']\n",
      "One Size Fits All: Context Collapse, Self-Presentation Strategies and Language Styles on Facebook ['size', 'fit', 'context', 'collapse', 'selfpresentation', 'strategie', 'language', 'style', 'facebook']\n",
      "Gender Prediction in Gender Segregated Societies Using Cell Phone Data ['gender', 'prediction', 'gender', 'segregated', 'societie', 'cell', 'phone', 'data']\n",
      "Network Characterization of Cartels in Public Auction Markets ['network', 'characterization', 'cartel', 'public', 'auction', 'market']\n",
      "Cultural Dynamics at Spotify ['cultural', 'dynamic', 'spotify']\n",
      "Personal Bests as Reference Points ['personal', 'reference', 'point']\n",
      "How many of your friends have access to the internet? Estimating internet adoption around the world from a sample of Facebook users ['friend', 'acce', 'internet', 'estimating', 'internet', 'adoption', 'world', 'sample', 'facebook', 'user']\n",
      "Even Good Bots Fight ['good', 'bot', 'fight']\n",
      "Ultrasound Technology and ‘Missing Women’ in India: Analyses and Now-casts Based on Google Searches ['ultrasound', 'technology', '‘missing', 'women’', 'india', 'analyse', 'nowcast', 'based', 'google', 'searche']\n",
      "Using Facebook Ads Audiences for Global Lifestyle Disease Surveillance ['facebook', 'ad', 'audience', 'global', 'lifestyle', 'disease', 'surveillance']\n",
      "The Effects of Conversation Threading on Online Discussion Networks ['effect', 'conversation', 'threading', 'online', 'discussion', 'network']\n",
      "Efficient Markov chain Monte Carlo Estimation of Exponential Random Graph Models ['efficient', 'markov', 'chain', 'monte', 'carlo', 'estimation', 'exponential', 'random', 'graph', 'model']\n",
      "Explaining protest intensity: The effect of roles and communities in a global cooperation network ['explaining', 'protest', 'intensity', 'effect', 'role', 'communitie', 'global', 'cooperation', 'network']\n",
      "Quantifying Attention to Electoral Integrity of Foreign States with Text Analysis ['quantifying', 'attention', 'electoral', 'integrity', 'foreign', 'state', 'text', 'analysis']\n",
      "Reputation and Success in Art ['reputation', 'succe', 'art']\n",
      "Stochastic popularity of social-media items ['stochastic', 'popularity', 'socialmedia', 'item']\n",
      "Assessing Human Error Against a Benchmark of Perfection ['assessing', 'human', 'error', 'benchmark', 'perfection']\n",
      "Fame and Obsolescence: Disentangling growth and ageing dynamics of patent citations ['fame', 'obsolescence', 'disentangling', 'growth', 'ageing', 'dynamic', 'patent', 'citation']\n",
      "Measuring attitudes to ethnic groups in the Russian-language social media ['measuring', 'attitude', 'ethnic', 'group', 'russianlanguage', 'social', 'media']\n",
      "Global warming is different from Racism: on the structure of controversy in Wikipedia ['global', 'warming', 'racism', 'structure', 'controversy', 'wikipedia']\n",
      "A Cross-Country Comparison of Crowdworker Motivations ['crosscountry', 'comparison', 'crowdworker', 'motivation']\n",
      "All Roads Lead to Lagos: Road networks and the legacies of extractive colonialism ['road', 'lead', 'lago', 'road', 'network', 'legacie', 'extractive', 'colonialism']\n",
      "Computational Affective Model of Human Action in a Social Dilemma ['computational', 'affective', 'model', 'human', 'action', 'social', 'dilemma']\n",
      "Patients’ written reviews as a resource for public healthcare management in England ['patients’', 'written', 'review', 'resource', 'public', 'healthcare', 'management', 'england']\n",
      "Economic Miracle, Political Disaster? Political and Social Consequences of the Hartz-Reforms (anonymized) ['economic', 'miracle', 'political', 'disaster', 'political', 'social', 'consequence', 'hartzreform', '(anonymized)']\n",
      "Making Money: Evidence from Individually Matched Network, Education, and Income Data ['making', 'money', 'evidence', 'individually', 'matched', 'network', 'education', 'income', 'data']\n",
      "The Evolution and Consequences of Peer Producing Wikipedia’s Rules ['evolution', 'consequence', 'peer', 'producing', 'wikipedia’', 'rule']\n",
      "Mapping the Higher Education System ['mapping', 'higher', 'education', 'system']\n",
      "To put that in perspective: Generating analogies that make numbers easier to understand ['put', 'perspective', 'generating', 'analogie', 'make', 'number', 'easier', 'understand']\n",
      "What do our digital records reveal about us? Automatic Prediction of Demographic and Psychometric Attributes from Digital Data ['digital', 'record', 'reveal', 'automatic', 'prediction', 'demographic', 'psychometric', 'attribute', 'digital', 'data']\n",
      "The Link Between Social Feedback and Weight Loss in a Reddit Community  ['link', 'social', 'feedback', 'weight', 'lo', 'reddit', 'community']\n",
      "Demographic trends and social dynamics of a financial bubble ['demographic', 'trend', 'social', 'dynamic', 'financial', 'bubble']\n",
      "Urban Scaling in US Presidential Elections ['urban', 'scaling', 'presidential', 'election']\n",
      "Are 140 Characters the Smoke Signal of the Digital Age? Using Twitter to Predict Community Suicide Rates ['140', 'character', 'smoke', 'signal', 'digital', 'age', 'twitter', 'predict', 'community', 'suicide', 'rate']\n",
      "Computing Quantitative Taxonomies of Social Interactions from Data ['computing', 'quantitative', 'taxonomie', 'social', 'interaction', 'data']\n",
      "Dyadic and network learning in online drug marketplaces: commitment or market efficiency? ['dyadic', 'network', 'learning', 'online', 'drug', 'marketplace', 'commitment', 'market', 'efficiency']\n",
      "Children names – simulations and facts ['children', '–', 'simulation', 'fact']\n",
      "Measuring the Value of Low-Cost Activism for Online Social and Political Movements ['measuring', 'lowcost', 'activism', 'online', 'social', 'political', 'movement']\n",
      "Online Network Organization of an Emergent Movement-Party ['online', 'network', 'organization', 'emergent', 'movementparty']\n",
      "Opportunities and Challenges in Computational Social Science: A Publisher’s Perspective ['opportunitie', 'challenge', 'computational', 'social', 'science', 'publisher’', 'perspective']\n",
      "Incentivized Data Sharing via Group-level Privacy-preservation ['incentivized', 'data', 'sharing', 'grouplevel', 'privacypreservation']\n",
      "Automatic Extraction of Protest Crowd Size from Newspaper Stories. ['automatic', 'extraction', 'protest', 'crowd', 'size', 'newspaper', 'storie']\n",
      "Trust is Extendible: An Online Experiment on Airbnb’s User Population ['trust', 'extendible', 'online', 'experiment', 'airbnb’', 'user', 'population']\n",
      "Studying Content Survival, Authorship & Controversy – By Tracing Every Word Change on Wikipedia ['studying', 'content', 'survival', 'authorship', '&', 'controversy', '–', 'tracing', 'word', 'change', 'wikipedia']\n",
      "Revealing the Anatomy of Vote Trading ['revealing', 'anatomy', 'vote', 'trading']\n",
      "Computational Approaches Toward Integrating Quantified Self Sensing and Social Media ['computational', 'approache', 'integrating', 'quantified', 'sensing', 'social', 'media']\n",
      "On the estimation of pairwise influences in dynamic choices ['estimation', 'pairwise', 'influence', 'dynamic', 'choice']\n",
      "Evolution of Moral Politics: Analysis on the Moral Foundations of Korean Political Parties ['evolution', 'moral', 'politic', 'analysis', 'moral', 'foundation', 'korean', 'political', 'partie']\n",
      "Small cities face greater impact from automation ['small', 'citie', 'face', 'greater', 'impact', 'automation']\n",
      "Looking for the strategies in the repeated prisoner's dilemma when the cooperation is established ['strategie', 'repeated', 'prisoner', 'dilemma', 'cooperation', 'established']\n",
      "Observing User Behavior in Group Recommender Systems ['observing', 'user', 'behavior', 'group', 'recommender', 'system']\n",
      "Do news recommenders foster filter bubbles? An empirical assessment of multiple recommender systems and their impact on content diversity ['recommender', 'foster', 'filter', 'bubble', 'empirical', 'assessment', 'multiple', 'recommender', 'system', 'impact', 'content', 'diversity']\n",
      "Hot issue publics on Twitter: Analyzing Charlotte protests with automated text analysis ['hot', 'issue', 'public', 'twitter', 'analyzing', 'charlotte', 'protest', 'automated', 'text', 'analysis']\n",
      "The Ikea-effect in collective problem solving ['ikeaeffect', 'collective', 'problem', 'solving']\n",
      "Spatial distributions of linguistic features are determined by the interaction of endogenous and exogenous dynamics ['spatial', 'distribution', 'linguistic', 'feature', 'determined', 'interaction', 'endogenou', 'exogenou', 'dynamic']\n",
      "Social Bayesian Learning in the Wisdom of the Crowd ['social', 'bayesian', 'learning', 'wisdom', 'crowd']\n",
      "Mapping out the social web from Big Data: Modeling selection of communication channels ['mapping', 'social', 'web', 'big', 'data', 'modeling', 'selection', 'communication', 'channel']\n",
      "Leveraging Large-Scale Ubiquitous Data for Fine-Grained Crime Prediction ['leveraging', 'largescale', 'ubiquitous', 'data', 'finegrained', 'crime', 'prediction']\n",
      "Women and Creativity: Gender Differences in the Production of Popular Music ['women', 'creativity', 'gender', 'difference', 'production', 'popular', 'music']\n",
      "Nature of Real and Perceived Bias in Chilean Media ['nature', 'real', 'perceived', 'bias', 'chilean', 'media']\n",
      "Cultivating the Appearance of Neutrality: Autocratic Propaganda in Africa and Asia ['cultivating', 'appearance', 'neutrality', 'autocratic', 'propaganda', 'africa', 'asia']\n",
      "A trigger-conditional Hawkes process model for social protest ['triggerconditional', 'hawke', 'process', 'model', 'social', 'protest']\n",
      "Propaganda and Protest: Evidence from Post-Cold War Africa ['propaganda', 'protest', 'evidence', 'postcold', 'war', 'africa']\n",
      "Differential Network Effects on Economic Outcomes: A Structural Perspective ['differential', 'network', 'effect', 'economic', 'outcome', 'structural', 'perspective']\n",
      "What comes first? Social strength or common friends? ['social', 'strength', 'common', 'friend']\n",
      "Turning Tweets into News: An Analysis of the Traditional Media Reach of Modi’s Twitter Feed ['turning', 'tweet', 'analysis', 'traditional', 'media', 'reach', 'modi’', 'twitter', 'feed']\n",
      "How Much Richer Do the Rich Get? The Matthew Effect and Non-Linear Preferential Attachment in Computer Programming ['richer', 'rich', 'matthew', 'effect', 'nonlinear', 'preferential', 'attachment', 'computer', 'programming']\n",
      "Problem-centric Network Embeddedness and Productivity in Open Source Software Communities ['problemcentric', 'network', 'embeddedne', 'productivity', 'open', 'source', 'software', 'communitie']\n"
     ]
    }
   ],
   "source": [
    "# now check what that looks like against our titles \n",
    "for talk in unique_talks:\n",
    "    title = talk[0][0]\n",
    "    keywords = get_keywords(title)\n",
    "    print(title, keywords) # looks pretty good"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# lets get a uniqe set of keywords, that we will use to populate our graph with. \n",
    "# if we use the defaultdict, we can find weights for our keywords \n",
    "unique_keywords = defaultdict(int)\n",
    "for talk in unique_talks:\n",
    "    title = talk[0][0]\n",
    "    keywords = get_keywords(title)\n",
    "    for keyword in keywords:\n",
    "        unique_keywords[keyword] += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2 digital 6\n",
      "8 evolution 5\n",
      "10 online 12\n",
      "12 twitter 11\n",
      "13 data 14\n",
      "16 economic 4\n",
      "17 cultural 5\n",
      "19 modeling 5\n",
      "22 social 26\n",
      "23 media 8\n",
      "26 record 3\n",
      "28 largescale 3\n",
      "33 market 5\n",
      "41 network 25\n",
      "42 dynamic 9\n",
      "43 influence 3\n",
      "44 wisdom 4\n",
      "45 crowd 5\n",
      "47 wikipedia 5\n",
      "53 moral 4\n",
      "54 foundation 3\n",
      "55 political 9\n",
      "58 analysis 9\n",
      "63 effect 8\n",
      "76 content 5\n",
      "83 measuring 3\n",
      "92 election 4\n",
      "103 text 3\n",
      "106 understanding 3\n",
      "107 collective 3\n",
      "110 learning 3\n",
      "130 evidence 3\n",
      "133 survey 3\n",
      "145 pattern 4\n",
      "156 complex 4\n",
      "167 computational 5\n",
      "175 global 4\n",
      "197 urban 3\n",
      "200 strategie 3\n",
      "216 gender 4\n",
      "219 community 3\n",
      "223 group 3\n",
      "227 bot 3\n",
      "235 propaganda 3\n",
      "276 model 5\n",
      "292 suicide 3\n",
      "301 difference 3\n",
      "318 facebook 4\n",
      "327 system 4\n",
      "332 cooperation 3\n",
      "354 mapping 3\n",
      "367 prediction 3\n",
      "374 public 3\n",
      "385 user 3\n",
      "411 protest 5\n",
      "475 perspective 3\n",
      "551 recommender 3\n"
     ]
    }
   ],
   "source": [
    "for key, value in enumerate(unique_keywords):\n",
    "    if unique_keywords[value] > 2:\n",
    "        print(key, value, unique_keywords[value])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "OK, now we have titles, authors and keywords, time to start thinking about creating a network of these kinds of objects to have a look at. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# connect to the database\n",
    "db = GraphDatabase(\"http://localhost:7474\", username=\"neo4j\", password=\"sage\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# we want three kinds of nodes - people, talks, keywords \n",
    "person = db.labels.create(\"Person\")\n",
    "talk = db.labels.create(\"Talk\")\n",
    "keyword = db.labels.create(\"Keyword\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "school 1\n",
      "segregation 1\n",
      "digital 6\n",
      "space 1\n",
      "bounded 1\n",
      "confidence 1\n",
      "extreme 1\n",
      "opinion 1\n",
      "evolution 5\n",
      " 1\n",
      "online 12\n",
      "experiment 2\n",
      "twitter 11\n",
      "data 14\n",
      "reflect 1\n",
      "link 2\n",
      "economic 4\n",
      "cultural 5\n",
      "modernization 1\n",
      "modeling 5\n",
      "echo 1\n",
      "chamber 1\n",
      "social 26\n",
      "media 8\n",
      "technique 1\n",
      "privacypreserving 1\n",
      "record 3\n",
      "linkage 1\n",
      "largescale 3\n",
      "science 2\n",
      "set 1\n",
      "ideological 1\n",
      "creation 1\n",
      "market 5\n",
      "socialist 1\n",
      "china 2\n",
      "change 2\n",
      "rhetoric 1\n",
      "people’ 1\n",
      "daily 1\n",
      "19462003 1\n",
      "network 25\n",
      "dynamic 9\n",
      "influence 3\n",
      "wisdom 4\n",
      "crowd 5\n",
      "read 1\n",
      "wikipedia 5\n",
      "theorybased 1\n",
      "targeting 1\n",
      "increase 1\n",
      "technology 2\n",
      "adoption 2\n",
      "moral 4\n",
      "foundation 3\n",
      "political 9\n",
      "discourse 2\n",
      "comparative 1\n",
      "analysis 9\n",
      "speech 2\n",
      "congre 1\n",
      "japanese 1\n",
      "diet 1\n",
      "effect 8\n",
      "missing 1\n",
      "translation 1\n",
      "analyse 2\n",
      "european 1\n",
      "legislative 1\n",
      "sampling 1\n",
      "attribute 2\n",
      "fall 1\n",
      "empire 1\n",
      "americanization 1\n",
      "english 1\n",
      "sponsored 1\n",
      "content 5\n",
      "psychological 2\n",
      "personality 1\n",
      "profile 2\n",
      "extremist 1\n",
      "care 1\n",
      "overhead 1\n",
      "measuring 3\n",
      "efficiency 2\n",
      "charity 1\n",
      "marketplace 2\n",
      "spread 1\n",
      "fake 1\n",
      "voter 1\n",
      "2016 2\n",
      "general 2\n",
      "election 4\n",
      "statistical 2\n",
      "scaling 2\n",
      "meet 1\n",
      "classic 1\n",
      "bayesian 2\n",
      "irtmodelling 1\n",
      "infer 1\n",
      "latent 1\n",
      "variable 1\n",
      "handcoded 1\n",
      "text 3\n",
      "memory 2\n",
      "remains\\\\ 1\n",
      "understanding 3\n",
      "collective 3\n",
      "age 2\n",
      "selective 1\n",
      "learning 3\n",
      "adaptive 1\n",
      "perceiving 1\n",
      "acting 1\n",
      "preference 1\n",
      "perception 1\n",
      "norm 1\n",
      "relate 1\n",
      "normative 1\n",
      "compliance 1\n",
      "named 1\n",
      "entitie 1\n",
      "computerautomated 1\n",
      "verbal 1\n",
      "deception 1\n",
      "detection 2\n",
      "inducing 1\n",
      "egocentric 1\n",
      "privacy 1\n",
      "setting 1\n",
      "evidence 3\n",
      "integrated 1\n",
      "representative 2\n",
      "survey 3\n",
      "exploring 1\n",
      "emotion 1\n",
      "movie 1\n",
      "review 2\n",
      "hostility 1\n",
      "bonu 1\n",
      "negativity 1\n",
      "dominate 1\n",
      "american 1\n",
      "communication 2\n",
      "inactivity 1\n",
      "pattern 4\n",
      "society 1\n",
      "promoting 1\n",
      "educational 1\n",
      "attainment 1\n",
      "scale 1\n",
      "intervention 1\n",
      "method 1\n",
      "acquiring 1\n",
      "analyzing 2\n",
      "visualizing 1\n",
      "complex 4\n",
      "street 1\n",
      "polarized 1\n",
      "skill 1\n",
      "complementarity 1\n",
      "transparency 1\n",
      "reuse 1\n",
      "open 2\n",
      "contest 1\n",
      "linking 1\n",
      "empirical 2\n",
      "computational 5\n",
      "approache 2\n",
      "identifying 1\n",
      "sink 1\n",
      "conduit 1\n",
      "offshore 1\n",
      "financial 2\n",
      "center 1\n",
      "global 4\n",
      "corporate 1\n",
      "ownership 1\n",
      "insurance 1\n",
      "fraud 1\n",
      "behavioral 1\n",
      "analytic 1\n",
      "approach 1\n",
      "ideology 1\n",
      "affective 2\n",
      "attitude 2\n",
      "mobility 2\n",
      "innovation 1\n",
      "comparing 1\n",
      "response 1\n",
      "visibility 1\n",
      "minoritie 2\n",
      "estimating 2\n",
      "possibility 1\n",
      "trajectorybased 1\n",
      "reidentification 1\n",
      "fusion 1\n",
      "urban 3\n",
      "dataset 1\n",
      "sharing 2\n",
      "strategie 3\n",
      "optimal 1\n",
      "team 1\n",
      "collaboration 1\n",
      "problem 2\n",
      "solving 2\n",
      "properties 1\n",
      "fluctuation 2\n",
      "time 1\n",
      "serie 1\n",
      "representing 1\n",
      "appearance 2\n",
      "word 2\n",
      "nationwide 1\n",
      "blog 1\n",
      "application 1\n",
      "gender 4\n",
      "outcome 2\n",
      "disparitie 1\n",
      "community 3\n",
      "designer 1\n",
      "modelling 1\n",
      "small 2\n",
      "group 3\n",
      "competing 1\n",
      "cooperating 1\n",
      "member 1\n",
      "bot 3\n",
      "japan’ 1\n",
      "2014 1\n",
      "preliminary 1\n",
      "result 1\n",
      "corpuslinguistic 1\n",
      "qualitative 1\n",
      "study 2\n",
      "propaganda 3\n",
      "large 1\n",
      "tagging 1\n",
      "ecosystem 1\n",
      "agent 1\n",
      "based 2\n",
      "interpret 1\n",
      "underlying 1\n",
      "factor 1\n",
      "underrepresentation 1\n",
      "hollywood 1\n",
      "film 1\n",
      "migration 2\n",
      "corridor 1\n",
      "cluster 1\n",
      "google+ 1\n",
      "studie 2\n",
      "linguistic 2\n",
      "neighbourhood 1\n",
      "explaining 2\n",
      "border 1\n",
      "multilingual 1\n",
      "coediting 1\n",
      "activity 1\n",
      "consumption 1\n",
      "traffic 1\n",
      "brexit 1\n",
      "campaign 2\n",
      "uk 2\n",
      "evolving 2\n",
      "affiliation 1\n",
      "politician 1\n",
      "quantifying 2\n",
      "research 1\n",
      "interest 1\n",
      "labvanced 1\n",
      "unified 1\n",
      "javascript 1\n",
      "framework 1\n",
      "conversation 2\n",
      "topicsentiment 1\n",
      "model 5\n",
      "analyze 1\n",
      "chinese 1\n",
      "national 1\n",
      "image 1\n",
      "intermodular 1\n",
      "connectivity 1\n",
      "create 1\n",
      "structural 2\n",
      "instability 1\n",
      "contagion 1\n",
      "attention 2\n",
      "promoted 1\n",
      "detecting 1\n",
      "russian 1\n",
      "celebrity 2\n",
      "suicide 3\n",
      "werther 1\n",
      "34 1\n",
      "oecd 1\n",
      "countrie 1\n",
      "wikipedia’ 2\n",
      "temporal 1\n",
      "correlation 1\n",
      "multiplex 1\n",
      "difference 3\n",
      "doe 1\n",
      "make 2\n",
      "languagebased 1\n",
      "heterogeneity 1\n",
      "firm 1\n",
      "performance 1\n",
      "uncertainty 2\n",
      "structure 2\n",
      "transaction 1\n",
      "sequence 1\n",
      "eu 1\n",
      "interbank 1\n",
      "money 2\n",
      "platform 1\n",
      "topic 1\n",
      "salience 1\n",
      "facebook 4\n",
      "target 1\n",
      "hub 1\n",
      "strategic 1\n",
      "diffusion 1\n",
      "regional 1\n",
      "economy 1\n",
      "decomposing 1\n",
      "labor 1\n",
      "system 4\n",
      "order 1\n",
      "law 1\n",
      "reputation 2\n",
      "promote 1\n",
      "cooperation 3\n",
      "cryptomarket 1\n",
      "illegal 1\n",
      "drug 2\n",
      "north 1\n",
      "korea 1\n",
      "view 1\n",
      "applying 1\n",
      "sentiment 1\n",
      "korean 2\n",
      "central 1\n",
      "agency 1\n",
      "report 1\n",
      "trust 2\n",
      "topology 1\n",
      "measurement 1\n",
      "facetobmi 1\n",
      "computer 2\n",
      "vision 1\n",
      "body 1\n",
      "weight 2\n",
      "picture 1\n",
      "mapping 3\n",
      "engineering 1\n",
      "design 1\n",
      "project 1\n",
      "email 1\n",
      "mining 1\n",
      "size 2\n",
      "fit 1\n",
      "context 1\n",
      "collapse 1\n",
      "selfpresentation 1\n",
      "language 1\n",
      "style 1\n",
      "prediction 3\n",
      "segregated 1\n",
      "societie 1\n",
      "cell 1\n",
      "phone 1\n",
      "characterization 1\n",
      "cartel 1\n",
      "public 3\n",
      "auction 1\n",
      "spotify 1\n",
      "personal 1\n",
      "reference 1\n",
      "point 1\n",
      "friend 2\n",
      "acce 1\n",
      "internet 2\n",
      "world 1\n",
      "sample 1\n",
      "user 3\n",
      "good 1\n",
      "fight 1\n",
      "ultrasound 1\n",
      "‘missing 1\n",
      "women’ 1\n",
      "india 1\n",
      "nowcast 1\n",
      "google 1\n",
      "searche 1\n",
      "ad 1\n",
      "audience 1\n",
      "lifestyle 1\n",
      "disease 1\n",
      "surveillance 1\n",
      "threading 1\n",
      "discussion 1\n",
      "efficient 1\n",
      "markov 1\n",
      "chain 1\n",
      "monte 1\n",
      "carlo 1\n",
      "estimation 2\n",
      "exponential 1\n",
      "random 1\n",
      "graph 1\n",
      "protest 5\n",
      "intensity 1\n",
      "role 1\n",
      "communitie 2\n",
      "electoral 1\n",
      "integrity 1\n",
      "foreign 1\n",
      "state 1\n",
      "succe 1\n",
      "art 1\n",
      "stochastic 1\n",
      "popularity 1\n",
      "socialmedia 1\n",
      "item 1\n",
      "assessing 1\n",
      "human 2\n",
      "error 1\n",
      "benchmark 1\n",
      "perfection 1\n",
      "fame 1\n",
      "obsolescence 1\n",
      "disentangling 1\n",
      "growth 1\n",
      "ageing 1\n",
      "patent 1\n",
      "citation 1\n",
      "ethnic 1\n",
      "russianlanguage 1\n",
      "warming 1\n",
      "racism 1\n",
      "controversy 2\n",
      "crosscountry 1\n",
      "comparison 1\n",
      "crowdworker 1\n",
      "motivation 1\n",
      "road 2\n",
      "lead 1\n",
      "lago 1\n",
      "legacie 1\n",
      "extractive 1\n",
      "colonialism 1\n",
      "action 1\n",
      "dilemma 2\n",
      "patients’ 1\n",
      "written 1\n",
      "resource 1\n",
      "healthcare 1\n",
      "management 1\n",
      "england 1\n",
      "miracle 1\n",
      "disaster 1\n",
      "consequence 2\n",
      "hartzreform 1\n",
      "(anonymized) 1\n",
      "making 1\n",
      "individually 1\n",
      "matched 1\n",
      "education 2\n",
      "income 1\n",
      "peer 1\n",
      "producing 1\n",
      "rule 1\n",
      "higher 1\n",
      "put 1\n",
      "perspective 3\n",
      "generating 1\n",
      "analogie 1\n",
      "number 1\n",
      "easier 1\n",
      "understand 1\n",
      "reveal 1\n",
      "automatic 2\n",
      "demographic 2\n",
      "psychometric 1\n",
      "feedback 1\n",
      "lo 1\n",
      "reddit 1\n",
      "trend 1\n",
      "bubble 2\n",
      "presidential 1\n",
      "140 1\n",
      "character 1\n",
      "smoke 1\n",
      "signal 1\n",
      "predict 1\n",
      "rate 1\n",
      "computing 1\n",
      "quantitative 1\n",
      "taxonomie 1\n",
      "interaction 2\n",
      "dyadic 1\n",
      "commitment 1\n",
      "children 1\n",
      "– 2\n",
      "simulation 1\n",
      "fact 1\n",
      "lowcost 1\n",
      "activism 1\n",
      "movement 1\n",
      "organization 1\n",
      "emergent 1\n",
      "movementparty 1\n",
      "opportunitie 1\n",
      "challenge 1\n",
      "publisher’ 1\n",
      "incentivized 1\n",
      "grouplevel 1\n",
      "privacypreservation 1\n",
      "extraction 1\n",
      "newspaper 1\n",
      "storie 1\n",
      "extendible 1\n",
      "airbnb’ 1\n",
      "population 1\n",
      "studying 1\n",
      "survival 1\n",
      "authorship 1\n",
      "& 1\n",
      "tracing 1\n",
      "revealing 1\n",
      "anatomy 1\n",
      "vote 1\n",
      "trading 1\n",
      "integrating 1\n",
      "quantified 1\n",
      "sensing 1\n",
      "pairwise 1\n",
      "choice 1\n",
      "politic 1\n",
      "partie 1\n",
      "citie 1\n",
      "face 1\n",
      "greater 1\n",
      "impact 2\n",
      "automation 1\n",
      "repeated 1\n",
      "prisoner 1\n",
      "established 1\n",
      "observing 1\n",
      "behavior 1\n",
      "recommender 3\n",
      "foster 1\n",
      "filter 1\n",
      "assessment 1\n",
      "multiple 1\n",
      "diversity 1\n",
      "hot 1\n",
      "issue 1\n",
      "charlotte 1\n",
      "automated 1\n",
      "ikeaeffect 1\n",
      "spatial 1\n",
      "distribution 1\n",
      "feature 1\n",
      "determined 1\n",
      "endogenou 1\n",
      "exogenou 1\n",
      "web 1\n",
      "big 1\n",
      "selection 1\n",
      "channel 1\n",
      "leveraging 1\n",
      "ubiquitous 1\n",
      "finegrained 1\n",
      "crime 1\n",
      "women 1\n",
      "creativity 1\n",
      "production 1\n",
      "popular 1\n",
      "music 1\n",
      "nature 1\n",
      "real 1\n",
      "perceived 1\n",
      "bias 1\n",
      "chilean 1\n",
      "cultivating 1\n",
      "neutrality 1\n",
      "autocratic 1\n",
      "africa 2\n",
      "asia 1\n",
      "triggerconditional 1\n",
      "hawke 1\n",
      "process 1\n",
      "postcold 1\n",
      "war 1\n",
      "differential 1\n",
      "strength 1\n",
      "common 1\n",
      "turning 1\n",
      "tweet 1\n",
      "traditional 1\n",
      "reach 1\n",
      "modi’ 1\n",
      "feed 1\n",
      "richer 1\n",
      "rich 1\n",
      "matthew 1\n",
      "nonlinear 1\n",
      "preferential 1\n",
      "attachment 1\n",
      "programming 1\n",
      "problemcentric 1\n",
      "embeddedne 1\n",
      "productivity 1\n",
      "source 1\n",
      "software 1\n"
     ]
    }
   ],
   "source": [
    "# we want to create a set of node objects for keywords and then add them to neo4j, with weights \n",
    "keyword_node_map = {}\n",
    "# for key, value in enumerate(unique_keywords):\n",
    "#     kw = value \n",
    "#     weight = unique_keywords[kw]\n",
    "#     print(kw, weight)\n",
    "#     keyword_node_map[kw] = db.nodes.create(name=kw, weight=weight)\n",
    "#     node = keyword_node_map[kw]\n",
    "#     keyword.add(node)\n",
    "\n",
    "# OK we have the nodes in with weights, I don't want to push them in again, so I'm going to comment out this code "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# we want to create a set of node objects for authors and then add them to neo4j, we don't need weights for these people\n",
    "unique_authors = []\n",
    "for talk in unique_talks:\n",
    "    for author in talk[1]:\n",
    "        if author.strip() not in unique_authors:\n",
    "            unique_authors.append(author.strip())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# # now generate nodes for our authors \n",
    "# author_node_map = {} \n",
    "# for author in unique_authors:\n",
    "#     author_node_map[author] = db.nodes.create(name=author)\n",
    "#     person.add(author_node_map[author])\n",
    "\n",
    "# OK got those in "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# now generate a set of nodes for the talks\n",
    "# talk = db.labels.create(\"Talk\")\n",
    "# titles = list(map(lambda x: x[0][0], unique_talks))\n",
    "# titles\n",
    "# talk_node_map = {}\n",
    "# for title in titles:\n",
    "#     talk_node_map[title] = db.nodes.create(name=title)\n",
    "#     talk.add(talk_node_map[title])\n",
    "\n",
    "Those are in now"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'talk_node_map' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-2-6df88738cd4f>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;31m# author_node_map\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0;31m# keyword_node_map\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mtalk_node_map\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'talk_node_map' is not defined"
     ]
    }
   ],
   "source": [
    "# author_node_map\n",
    "# keyword_node_map\n",
    "talk_node_map"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "ow we want to create the links between the nodes, there are a number of different links we want to make. \n",
    "\n",
    "author - author links \n",
    "author - keyword links \n",
    "talk - author links\n",
    "\n",
    "let's start with the talk - author links "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/991>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/618>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/992>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/619>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/620>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/621>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/622>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/623>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/624>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/625>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/993>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/626>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/627>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/994>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/628>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/629>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/630>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/631>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/995>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/632>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/633>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/996>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/634>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/635>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/997>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/636>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/637>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/638>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/998>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/639>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/640>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/641>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/642>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/643>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/644>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/645>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/999>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/646>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/647>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/648>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/649>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1000>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/650>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/651>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1001>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/652>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/653>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/654>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1002>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/655>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/640>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/656>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/657>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/644>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1003>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/658>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/659>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/660>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/661>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1004>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/662>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/663>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/664>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/665>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/666>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1005>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/667>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/668>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/669>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/670>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/671>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1006>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/672>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/673>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1007>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/674>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/675>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/676>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/677>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/678>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/679>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/680>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/681>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1008>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/682>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/683>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1009>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/684>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/685>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/686>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/687>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1010>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/688>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/689>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/690>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/691>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/692>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/693>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1011>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/694>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/695>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/696>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/697>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1012>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/698>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/699>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/700>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/701>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1013>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/702>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/703>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1014>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/704>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/705>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/706>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/707>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1015>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/708>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/709>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/710>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1016>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/711>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/712>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/713>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/714>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1017>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/715>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1018>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/716>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1019>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/717>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/718>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/719>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/720>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/721>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/722>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1020>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/723>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/724>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1021>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/725>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/726>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/727>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/728>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1022>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/729>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/730>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/731>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/732>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1023>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/733>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/734>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/735>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1024>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/656>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/736>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/655>\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<Neo4j Node: http://localhost:7474/db/data/node/640>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/644>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1025>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/737>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/738>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/739>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/740>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1026>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/741>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/742>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/743>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/744>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/745>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1027>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/746>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/747>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/748>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/749>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1028>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/750>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/751>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/752>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/753>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1029>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/754>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/755>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1030>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/756>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/757>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/758>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1031>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/759>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/760>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/761>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1032>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/762>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1033>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/763>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/764>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/668>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/765>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1034>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/766>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/656>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/767>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/768>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/644>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1035>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/769>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/770>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/771>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/772>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/773>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/774>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1036>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/773>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/775>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/771>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/772>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/770>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1037>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/776>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/777>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/778>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1038>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/779>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/780>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/781>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/782>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/783>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1039>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/784>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/628>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1040>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/785>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/786>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/787>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/788>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1041>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/789>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/790>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1042>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/791>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/792>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/793>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/794>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1043>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/795>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/796>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/797>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1044>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/798>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/799>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1045>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/800>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/801>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/802>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1046>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/803>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/697>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/696>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1047>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/804>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1048>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/805>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/806>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/807>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/644>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1049>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/808>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/809>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/810>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1050>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/719>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/717>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/811>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/722>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1051>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/812>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/813>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/814>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1052>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/815>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/816>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1053>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/817>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/818>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/819>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/820>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/821>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1054>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/822>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/823>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/668>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/824>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/825>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/826>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/827>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1055>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/828>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/829>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/830>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/831>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1056>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/832>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/833>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/834>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/835>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/836>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/837>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1057>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/838>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/839>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/840>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1058>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/750>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/841>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1059>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/842>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/843>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/844>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1060>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/845>\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<Neo4j Node: http://localhost:7474/db/data/node/846>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1061>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/847>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/848>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1062>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/686>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/684>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/849>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/687>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1063>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/850>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/851>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/852>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/853>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/668>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1064>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/854>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/855>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/668>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/764>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1065>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/856>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/793>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/857>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1066>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/858>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/859>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/860>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/861>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/862>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1067>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/863>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1068>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/864>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/865>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/866>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1069>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/867>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/868>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/869>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/870>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1070>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/871>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1071>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/845>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/872>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/873>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1072>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/874>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/875>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/876>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/877>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1073>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/878>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/879>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/880>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/881>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/882>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1074>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/883>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/862>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1075>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/884>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/806>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/885>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/644>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1076>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/886>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/887>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/888>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/889>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/681>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1077>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/890>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/891>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1078>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/892>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1079>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/893>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1080>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/894>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/895>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/896>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/897>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/898>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1081>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/899>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1082>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/900>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/901>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1083>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/902>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/903>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/904>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1084>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/905>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/906>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/907>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/908>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1085>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/909>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/668>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/910>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1086>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/911>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/912>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/913>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1087>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/914>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/915>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/916>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1088>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/917>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/918>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1089>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/919>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1090>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/813>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/814>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/920>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1091>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/921>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/922>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1092>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/923>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/672>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1093>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/856>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/794>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/792>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/793>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1094>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/924>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/925>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1095>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/926>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/927>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1096>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/928>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/929>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1097>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/930>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/931>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/932>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/933>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1098>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/885>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/934>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/935>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1099>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/936>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/937>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1100>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/938>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/939>\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<Neo4j Node: http://localhost:7474/db/data/node/668>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1101>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/940>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1102>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/941>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1103>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/717>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/719>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/942>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/811>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/722>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1104>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/943>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/944>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1105>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/945>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1106>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/946>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/947>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/948>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/949>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1107>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/950>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/951>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/952>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/953>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1108>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/954>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/955>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/956>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1109>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/957>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/958>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/959>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1110>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/960>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/961>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/962>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/690>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/693>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1111>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/963>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/964>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/965>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/966>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/967>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1112>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/968>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/969>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1113>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/970>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/971>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/972>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/973>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/745>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1114>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/974>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/975>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/976>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1115>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/977>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/978>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1116>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/979>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1117>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/978>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/977>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1118>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/894>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/980>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/981>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/897>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/982>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/983>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/984>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1119>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/982>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/985>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/942>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1120>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/928>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/986>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/987>\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/988>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1121>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/989>\n",
      "talk:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/1122>\n",
      "authors:\n",
      "<Neo4j Node: http://localhost:7474/db/data/node/990>\n"
     ]
    }
   ],
   "source": [
    "# create links between talks and authors \n",
    "# for talk in unique_talks:\n",
    "#     title = talk[0][0]\n",
    "#     authors = talk[1]\n",
    "#     talk_node_id = talk_node_map[title]\n",
    "#     print(\"talk:\")\n",
    "#     print(talk_node_id)\n",
    "#     print(\"authors:\")\n",
    "#     for author in authors:\n",
    "#         author_node_id = author_node_map[author.strip()]\n",
    "#         print(author_node_id)\n",
    "#         author_node_id.Wrote(talk_node_id)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 203,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def gen_co_authorships(author_list):\n",
    "    \"\"\"\n",
    "    return a set of tuples of all co-authorships \n",
    "    \"\"\"\n",
    "    pairs = []\n",
    "    for author in author_list:\n",
    "        for target in author_list:\n",
    "            a = author.strip()\n",
    "            b = target.strip()\n",
    "            pair = sorted([a, b])\n",
    "            if pair not in pairs and a != b:\n",
    "                pairs.append(pair)\n",
    "    return pairs "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Samuel Martin', ' Sylvie Huet', ' Pascal Gend', ' Mohammed Amine Ait Oumajoud', ' Guillaume Deffuant', ' Armelle Nugier', 'Serge Guimond']\n",
      "['Jochen Hirschle', 'Tuuli-Marja Kleiner']\n",
      "['Kazutoshi Sasahara', ' Giovanni Luca Ciampaglia', ' Alessandro Flammini', 'Filippo Menczer']\n",
      "['Rainer Schnell', 'Christian Borgs']\n",
      "['Shilin Jia', 'Linzhuo Li']\n",
      "['Joshua Becker', ' Devon Brackbill', 'Damon Centola']\n",
      "['Florian Lemmerich', ' Philipp Singer', ' Robert West', ' Leila Zia', ' Ellery Wulczyn', ' Markus Strohmaier', 'Jure Leskovec']\n",
      "['Lori Beaman', ' Ariel Benyishay', ' Jeremy Magruder', 'Ahmed Mushfiq Mobarak']\n",
      "['Hiroki Takikawa', 'Takuto Sakamoto']\n",
      "['Laura Hollink', ' Astrid van Aggelen', 'Jacco van Ossenbruggen']\n",
      "['Claudia Wagner', ' Philipp Singer', ' Fariba Karimi', ' Jürgen Pfeffer', 'Markus Strohmaier']\n",
      "['Bruno Gonçalves', ' Lucia Loureiro-Porto', ' Jose J. Ramasco', 'David Sanchez']\n",
      "['Jaimie Park', ' Mahmoudreza Babaei', ' Przemyslaw Grabowicz', ' Krishna Gummadi', 'Sue Moon']\n",
      "['Meysam Alizadeh', ' Ingmar Weber', ' Claudio Cioffi', ' Santo Fortunato', 'Michael Macy']\n",
      "['Ceren Budak', 'Justin Rao']\n",
      "['Kenneth Joseph', ' Nir Grinberg', ' Lisa Friedland', ' William Hobbs', ' Carolina Mattsson', ' Stefan McCabe', ' Lucas Da Silva Almeida', 'David Lazer']\n",
      "['Thomas Daeubler', 'Kenneth Benoit']\n",
      "['Ruth Garcia Gavilanes', ' Anders Mollgaard', ' Milena Tsvetkova', 'Taha Yasseri']\n",
      "['Alejandro Noriega Campero', ' Abdullah Almaatouq', ' Peter Krafft', ' Abdulrahman Alotaibi', ' Mehdi Moussaid', 'Alex Pentland']\n",
      "['Richard Lu', ' Jennifer Chatman', ' Sameer Srivastava', 'Amir Goldberg']\n",
      "['Bennett Kleinberg', ' Maximilian Mozes', ' Arnoud Arntz', 'Bruno Verschuere']\n",
      "['Yang-Chih Fu', 'Ming-Yi Chang']\n",
      "['Osnat Mokryn', ' Nadim Bader', ' Joel Lanir', 'Yael Albo']\n",
      "['Solomon Messing', ' Patrick Van Kessel', 'Adam Hughes']\n",
      "['David Garcia', ' Pavlin Mavrodiev', ' Daniele Casati', 'Frank Schweitzer']\n",
      "['Morgan Frank', ' Ahmad Alabdulkareem', ' Lijun Sun', ' Bedoor Alshebli', ' Cesar Hidalgo', 'Iyad Rahwan']\n",
      "['Sabine Brunswicker', 'Michael Prietula']\n",
      "['Javier Garcia-Bernardo', ' Jan Fichtner', ' Eelke Heemskerk', 'Frank Takes']\n",
      "['Eray Alpan', ' Gokhan Gokturk', ' Selim Balcisoy', 'Burcin Bozkaya']\n",
      "['Nikolas Zöller', ' Ingo Wolf', 'Tobias Schröder']\n",
      "['Fariba Karimi', ' Mathieu Genois', ' Claudia Wagner', ' Philipp Singer', 'Markus Strohmaier']\n",
      "['Dániel Kondor', ' Behrooz Hashemian', ' Yves-Alexandre de Montjoye', 'Carlo Ratti']\n",
      "['Pj Lamberson', ' John Lang', ' Noshir Contractor', ' Leslie Dechurch', 'Brian Uzzi']\n",
      "['Hayafumi Watanabe', ' Yukie Sano', ' Hideki Takayasu', 'Misako Takayasu']\n",
      "['Johannes Wachs', ' Aniko Hannak', ' Balint Daroczy', 'Andras Voros']\n",
      "['Ismo Koponen', 'Maija Nousiainen']\n",
      "['Stefan Evert', ' Philipp Heinrich', 'Fabian Schaefer']\n",
      "['Yasuhiro Hashimoto', ' Mizuki Oka', 'Takashi Ikegami']\n",
      "['Johnnatan Messias', ' Fabricio Benevenuto', ' Ingmar Weber', 'Emilio Zagheni']\n",
      "['Anna Samoilenko', ' Fariba Karimi', ' Daniel Edler', ' Jérôme Kunegis', 'Markus Strohmaier']\n",
      "['Lorien Jasny', ' Susan Banducci', ' Travis Coan', ' Iulia Cioroianu', ' Iain Weaver', 'Hywell Williams']\n",
      "['Iain Weaver', ' Hywel Williams', ' Travis Coan', ' Iulia Cioroianu', 'Susan Banducci']\n",
      "['Tao Jia', ' Dashun Wang', 'Boleslaw Szymanski']\n",
      "['Holger Finger', ' Caspar Goeke', ' Dorena Diekamp', ' Kai Standvoß', 'Peter König']\n",
      "['Rishemjit Kaur', 'Kazutoshi Sasahara']\n",
      "['Yuemei Xu', ' Jiayi Zhou', ' Xiaoya Zhang', 'Chenyu Dong']\n",
      "['Jana Huisman', 'Olivia Woolley']\n",
      "['Marijn ten Thij', ' David Laniado', ' Andreas Kaltenbrunner', 'Yana Volkovich']\n",
      "['Denis Stukal', ' Sergey Sanovich', 'Joshua A. Tucker']\n",
      "['Mark Lutter', 'Karlijn Roex']\n",
      "['Michele Starnini', ' Andrea Baronchelli', 'Romualdo Pastor-Satorras']\n",
      "['Matthew Corritore', ' Amir Goldberg', 'Sameer Srivastava']\n",
      "['Sebastian Stier', ' Arnim Bleier', ' Haiko Lietz', 'Markus Strohmaier']\n",
      "['Aamena Alshamsi', ' Flavio Pinheiro', 'César Hidalgo']\n",
      "['Lijun Sun', ' Morgan Frank', ' Hyejin Youn', 'Iyad Rahwan']\n",
      "['Wojtek Przepiorka', ' Lukas Norbutas', 'Rense Corten']\n",
      "['Weiqi Zhang', 'Dmitry Zinoviev']\n",
      "['Santa Agreste', ' Pasquale De Meo', ' Emilio Ferrara', ' Sebastiano Piccolo', 'Alessandro Provetti']\n",
      "['Enes Kocabey', ' Mustafa Camurcu', ' Ingmar Weber', ' Ferda Ofli', ' Yusuf Aytar', ' Javier Marin', 'Antonio Torralba']\n",
      "['Pedro Parraguez', ' Melanie Kreye', ' Phil Cash', 'Anja Maier']\n",
      "['Teresa Gil Lopez', ' Cuihua Shen', ' Grace Benefield', ' Nicholas Palomares', ' Michal Kosinski', 'David Stillwell']\n",
      "['Riyadh Alnasser', ' Faisal Aleissa', 'Michele Catasta']\n",
      "['Johannes Wachs', 'János Kertész']\n",
      "['Marc Keuschnigg', ' Martin Arvidsson', 'Peter Hedström']\n",
      "['Ashton Anderson', 'Etan Green']\n",
      "['Dennis Feehan', 'Curtiss Cobb']\n",
      "['Milena Tsvetkova', ' Ruth Garcia Gavilanes', ' Luciano Floridi', 'Taha Yasseri']\n",
      "['Ridhi Kashyap', ' Francesco Billari', ' Nicolo Cavalli', ' Eric Qian', 'Ingmar Weber']\n",
      "['Matheus Araújo', ' Yelena Mejova', ' Ingmar Weber', 'Fabricio Benevenuto']\n",
      "['Pablo Aragón', ' Andreas Kaltenbrunner', 'Vicenç Gómez']\n",
      "['Maksym Byshkin', ' Alex Stivala', ' Mira Antonietta', ' Garry Robins', 'Alessandro Lomi']\n",
      "['Ashrakat Elshehawy', ' Nikolay Marinov', 'Federico Nanni']\n",
      "['Samuel Fraiberger', ' Roberta Sinatra', ' Christoph Riedl', 'Laszlo Barabasi']\n",
      "['Ashton Anderson', ' Jon Kleinberg', 'Sendhil Mullainathan']\n",
      "['Kyle Higham', ' Michele Governale', ' Adam Jaffe', 'Ulrich Zülicke']\n",
      "['Koltsova Olessia', ' Nikolenko Sergey', ' Koltcov Sergei', ' Alexeeva Svetlana', 'Nagorny Oleg']\n",
      "['Juergen Lerner', 'Alessandro Lomi']\n",
      "['Lisa Posch', ' Arnim Bleier', ' Fabian Flöck', 'Markus Strohmaier']\n",
      "['Philipp Hunziker', ' Carl Muller-Crepon', ' Matthew Simonson', ' Lars-Erik Cederman', 'David Lazer']\n",
      "['Joshua Jung', 'Jesse Hoey']\n",
      "['Eaman Jahani', ' Guillaume Saint‑jacques', ' Pal Sundsøy', ' Johannes Bjelland', \"Alex 'Sandy' Pentland\"]\n",
      "['Flavio L. Pinheiro', 'Sara Encarnação']\n",
      "['Christopher Riederer', ' Jake Hofman', 'Daniel Goldstein']\n",
      "['Kyriaki Kalimeri', ' Mariano Gastón Beiró', ' Robert Raleigh', 'Ciro Cattuto']\n",
      "['Tiago Cunha', ' Ingmar Weber', 'Gisele Pappa']\n",
      "['Federico Musciotto', ' Jyrki Piilo', 'Rosario N. Mantegna']\n",
      "['Eszter Bokányi', ' Zoltán Szállási', 'Gábor Vattay']\n",
      "['Allison Phillips', 'Jing Zhao']\n",
      "['Lukas Norbutas', ' Rense Corten', 'Stijn Ruiter']\n",
      "['Krzysztof Kulakowski', 'Malgorzata Krawczyk']\n",
      "['Lia Bozarth', 'Ceren Budak']\n",
      "['Pablo Aragón', ' Yana Volkovich', ' David Laniado', 'Andreas Kaltenbrunner']\n",
      "['Katie Metzler', 'Ian Mulvany']\n",
      "['Stefano Bennati', 'Evangelos Pournaras']\n",
      "['Sunandan Chakraborty', 'Pablo Barberá']\n",
      "['Bruno Abrahao', ' Paolo Parigi', ' Alok Gupta', 'Karen Cook']\n",
      "['Fabian Flöck', ' Kenan Erdogan', 'Maribel Acosta']\n",
      "['Omar Guerrero', 'Ulrich Matter']\n",
      "['Munmun De Choudhury', ' Mrinal Kumar', 'Ingmar Weber']\n",
      "['Morgan Frank', ' Lijun Sun', ' Manuel Cebrian', ' Hyejin Youn', 'Iyad Rahwan']\n",
      "['Jelena Grujic', 'Tom Lenaerts']\n",
      "['Judith Moeller', ' Damian Trilling', ' Natali Helberger', 'Bram van Es']\n",
      "['Ryan Wesslen', ' Tiffany Gallicano', ' Jean-Claude Thill', 'Sagar Nandu']\n",
      "['Oana Vuculescu', ' Carsten Bergenholtz', 'Michela Beretta']\n",
      "['Henri Kauhanen', ' Deepthi Gopal', 'Tobias Galla']\n",
      "['Dhaval Adjodah', ' Yan Leng', ' Shi Kai Chong', ' Peter Krafft', 'Alex Pentland']\n",
      "['Janos Kertesz', ' Janos Török', ' Yohsuke Murase', ' Hang-Hyun Jo', 'Kimmo Kaski']\n",
      "['Cristina Kadar', 'Irena Pletikosa']\n",
      "['Michael Mauskapf', ' Jared Lorince', ' Noah Askin', ' Emoke-Agnes Horvat', 'Brian Uzzi']\n",
      "['Erick Elejalde', ' Leo Ferres', 'Eelco Herder']\n",
      "['Erin Carter', 'Brett Carter']\n",
      "['Brett Carter', 'Erin Carter']\n",
      "['Eaman Jahani', ' Guillaume Saint-Jacques', ' Pål Sundsøy', ' Johannes Bjelland', ' Esteban Moro', '', \"Alex 'Sandy’ Pentland\"]\n",
      "['Esteban Moro', ' Giovanna Miritello', 'Manuel Cebrian']\n",
      "['Sunandan Chakraborty', ' Joyojeet Pal', ' Priyank Chandra', 'Daniel Romero']\n"
     ]
    }
   ],
   "source": [
    "# create links between authors \n",
    "# we want to not create a link if it already exists. It might be interesting to look at relationship weights, but lets not \n",
    "# do that now. #\n",
    "\n",
    "# though we could check for existence of connections in the DB, the sad truth is that I don't know the query API well enough\n",
    "# yet, and have about 15 minutes to finish this before I go out an pick up some fish, so I have to do this the quick way\n",
    "author_author_map = [] \n",
    "global_connections = []\n",
    "for talk in unique_talks:\n",
    "    authors = talk[1]\n",
    "    if len(authors) > 1:\n",
    "        print(authors)\n",
    "        co_authorships = gen_co_authorships(authors)\n",
    "        for co_authorship in co_authorships:\n",
    "            if co_authorship not in global_connections:\n",
    "                global_connections.append(co_authorship)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# we have now generated all global connections betweeb just authors, we can create the links between these authors! \n",
    "for connection in global_connections:\n",
    "    source_node = author_node_map[connection[0]]\n",
    "    target_node = author_node_map[connection[1]]\n",
    "    source_node.CoAuthor(target_node)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Interesting Grail Queries against this graph\n",
    "\n",
    "MATCH (n:Keyword) WHERE n.weight > 4 RETURN n LIMIT 700"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}