YouTube Likes Count Prediction

This method uses various OSM platforms' data to predict likes on a YouTube video

Steps to run (Assuming Ubuntu 16.04 with python3.X and mongoDB installed and setup):

First install all the requirements:

$ pip3 install -r requirements.txt

Next import the raw.json and processed.json as :

$ mongoimport -d PreCog -c YoutubeRaw -h localhost:27017 raw.json
$ mongoimport -d PreCog -c YoutubeProcessed -h localhost:27017 processed.json

Now if you do not want to add the required API credentials, for testing purpose you can use the video IDS in 'testVideos.txt' via the Flask Web App.

$ python3 server.py

To run for any video add the following detials to the mentioned files :

• Ibm Tone Analyser Credentials - utils/comments_analyser.py
• Facebook Graph API access token - utils/facebook_search.py
• Google Account Details - utils/google_trends.py
• Reddit API details - utils/reddit_search.py
• YouTubeAPI developer KEY - utils/youtube_api.py

Finally run :

$ python3 server.py

You should now be able to use the flask webapp (http://localhost:5000/) to make predictions