We rate dog is a twitter account with funny and interesting tweets of rating dogs
Project objectives
The project main objectives were:
- Perform data wrangling (gathering, assessing and cleaning) on the provided sources of data.
- Store, analyze, and visualize the wrangled data.
- Reporting on
- data wrangling efforts.
- data analyses and visualizations.
In addition, as per project specificacion, only original tweets/ratings that have images should be used in the analysis (no retweets nor replies).
This project was completed as part of Udacity's Data Analysis Professional Nanodegree.
- twitter-archive-enhanced.csv provided from Udacity.
- image-predictions.tsv Downloaded from this link
- tweet_json.txt Gathered from twitter API.
- twitter_archive_master.db database hold the cleaned combined dataset.
- Wrangle report: documentation for data wrangling steps: gather, assess, and clean.
- Act report: documentation of analysis and insights into final data.
- Jupyter notebook
- Python and it's libraries (pandas, numpy, matplotlib, seaborn, requests, tweepy, worldcloud, io, os, PIL, json, sqlalchemy)
- Microsoft Office programms (Excel, Word, Power Point)
- Funny youtube video to know the difference between ('doggo', 'floofer', 'pupper', 'puppo')
- Reading pandas data frame row by row Stack over flow
- For the word cloud funny image I followed this tutorial at DataCamp
