Skip to content

Latest commit

 

History

History
52 lines (38 loc) · 2.15 KB

README.md

File metadata and controls

52 lines (38 loc) · 2.15 KB

Twitter user report

Open In Colab

Objective

The purpose of the project is to provide a quick overview about the activity and the engagement related to a twitter profile. The account analyzed is related to the italian politician Matteo Salvini because his profile is placed among the most active Twitter profiles in Italy.

Here's the pipeline

Pipeline

pipeline

Methods Used

  • Scraping: gather tweets with the library Twint. This library allows to collect almost the entire activity of a user without limitations
  • Data Profiling: quick transformations (casting and field drop/rename) made with Pandas and PdPipe
  • Data Visualization: made with Plotly and obtained the following plots:
    • Tables exposing:
      • All the collected tweets
      • Top 5 retweeted posts
      • Top 5 replied posts
      • Top 5 liked posts
    • Engagement overview over the time
      • Number of tweets progression
      • Number of likes progression
      • Numbers of retweets progression
      • Number of replies progression
    • Waffle chart to show the overall activity on weekday and month
  • Data Consumption: static report made with Datapane

Libraries

Name Link
Twint https://github.com/twintproject/twint
pdPipe https://pdpipe.github.io/pdpipe/
pysqldf https://pypi.org/project/pysqldf/
Seaborn https://seaborn.pydata.org/
Ploltly https://plotly.com/
Datapane https://datapane.com/

Results

DataPane

https://datapane.com/u/airaghidavide/reports/M7bQbw3/twitter-profile-report-matteo-salvini/

tables_tweets engagemnt heatmap