The world of data mining is full of endless opportunities for exploration. New apps for data gathering, analysis and visualization are being developed constantly and exponentially. Our inspiration was tying in this exciting field to the social environment. Technology can affect our social world, and this is the next step to influencing our generation which is so reliant on social media.
What it does
It takes in status and comment data from Facebook pages and applies our algorithms to return a sentiment value. Once the data is processed, visualizations are offered to the user to give them insights for their page.
How we built it
We used Python 2.7 for the data mining and machine learning. The data from there can then be accessed through a REST api call. Our front end is made using React which makes the API call and then using Chart.js, creates preliminary visualizations for the user.
Challenges we ran into
For many of us, this was our first time using our respective technologies (Python + Flask, Node + React, etc.) One particularly difficult challenge on the front end was the accessibility of the latest Charts.js library as a react component. We were forced to use an older library for support with our stack, which ended up restricting our options on how we could visualize our data.
Accomplishments that we're proud of
We're able to scrape and process megabytes of raw text data in seconds thanks to our linear run-time algorithm.
Fields we use (also available for download for personal analysis) :
( "status_id", "status_message", "link_name", "status_type", "status_link", "status_published", "num_reactions", "num_comments", "num_shares", "num_likes", "num_loves", "num_wows", "num_hahas", "num_sads", "num_angrys", "good_reception", "bad_reception", "reduce_message", "pos_words", "neg_words", "neu_words" )