Wrangle & Analyze Data

This project was completed as part of the course requirements of Udacity's Data Analyst Nanodegree certification.

Overview

The project used data from the WeRateDogs Twitter account. The data was assessed, cleaned and analyzed to provide accurate insights into account follower behaviour.

An online summary of the material can be found at my blog.

The project involved gathering data using a variety of file types and gathering techniques (manual download, programmatic download, api access), assessing the data for quality and tidiness, cleaning the data using a define, code, test methodology, and completing analysis and visualzations of the cleaned datasets.

Technologies Used

• Python
• Libraries: pandas, numpy, matplotlib, seaborn, json, os, requests, tweepy
• Jupyter Notebook

Key Findings

• The WeRateDogs account saw a decline in followers over approximately two years
• During the same period, total favorites received per week generally increased
• Age of the dog is less likely to influence numbers of retweets and favorites, but especially fluffy dogs receive more favorites on average
• While there is some correlation between the ratings provided by the site and corresponding retweets and favorites, there is still substantial variability across the rating spectrum