Wrangling @dog_rates (WeRateDogs) tweets and related data

Overview

Project consisted of gathering data from multiple disparate sources to be cleaned and combined into a single tidy dataset.

Required steps:

• Gather data from a variety of sources and in different formats

  • Import provided twitter-archive-enhanced.csv file from disk
  • Download the image-predictions.tsv file from Udacity using requets library
  • Call Twitter API to get full JSON objects for each tweet_id in the provided archive and store to a single txt file
  • Read in the generated tweet_json file from the API call

• Assess and fix quality and tidiness of data - Find issues w/ the data - Fix them

• Showcase/analyze the complete, clean, and tidy dataset

  • Analyze the dataset to find any interesting relationships
    • Ratings over time
    • Ratings by type/name (doggo, floofer, pupper, puppo)
    • What are the most common breed predictions
  • Visualize the findings