Hackathon_uO

Inspiration

The data that was given to us as well as additional data scrapped from Facebook and twitter for chinese accounts.

What it does

It shows the impact of Chinese accounts in different countries in term of likes, retweets, shares etc.. (interaction in social media)

How we built it

Python, Tableau, Excel

Challenges we ran into

Time mainly

Accomplishments that we're proud of

Having interesting insights that focuses on the current Israel-Hamas conflict

What we learned

We learned how to visualize text data

Story

Overall we want to see the Chinese propaganda on social media using the acronym SEE (scale, effort, effect).

Scale can be inferred using the impact of Chinese social media by looking at number of followers in different regions, number of countries impacted, languages they are using

Effort can be measured through the analogy where if a guy sleeps for eight hours, a single Chinese media account uploads 17h of video content

Effect can be seen through the number of viral posts in specific regions like EU and USA about important topics happening in real time (The israel-Hamas war)

Procedure

For normal visualization we used Tableau and Excel We checked facts on followers and influential users from the internet We tried mapping our whole analysis in the first steps with facts collected from the internet We scrapped additional data like tweets posts using technologies like selenium, puppeteer and in-house API We encountered road-block due to policy changes in Twitter especially We used apify to scrape the rest of the data: it gave us +3000 from Facebook and +2000 from twitter of different accounts We started cleaning the data narrowing down for example language to english, removing stop words and punctuations as well as creating bag of words We tried visualizing the data through wordcloud library : we were not able to gather at this step much information because the interference was subtle We tried capturing the amount of videos posted by these accounts : for one account for example there was at certain date a dump of posts (IGTN europe) so we started investigating the types of posts of this media outlets and found that the media’s main focus region was EU and they were posting a lot about the israel-hamas conflict. We started analyzing in-depth data that we scrapped from facebook. We also tried to see for viral videos especially ones related to the conflicts. We found out that those posts are mainly targeting USA and the EU and tried to create a word cloud of the words used in these posts. We used text to Ai