-
Scraping data from youtube
- Youtube API key is important in case you want to connect to youtube Dataset
- Use this key in your script to get data of trending videos on a particular day.
-
Cleaning Data
- Use lambda function to clean data
-
Model Building:
- Using youtube api key get data about recently added videos
- train classification model on trending data and recent data to classify whether video can be in trending tab or not
- need to run the script every day for at least two months to gather significant amount of data to train ML Model
- recently updated on 26/04/2020
- In order to predict whether a video can go viral or not create classification model whcih is trained on youtube data
- Hyper Parameter Tuning is required to selecet best possible algorithm
- Gradient Boosting, SVR, Random Forest, Decision Tree, MLP Classifier are amongst who performs well.
michelangelo367 / youtube-trending Goto Github PK
View Code? Open in Web Editor NEWThis project forked from tejan4422/youtube-trending