stefanodesaraca-YouTube-Videos-Data-Prediction-with-Machine-Learning

This project aims to: scrape videos data of a given YouTube channel using the YouTube Data API V3, execute a complete EDA and predict future videos' data. The project comprehends a list of options accessible from a menu which let the user:

1. Download the video's data of the chosen YouTube channel
2. Convert the JSON file containing the videos' data into CSV
3. Get a full EDA and Machine Learning analysis of the dataset

The EDA includes various insights on the data and a Shapiro-Wilk normality test to check if the distribution of the variables is a normal
The Machine Learning analysis includes:

1. Feature selection process using the "Forward" method implemented through a RandomForestRegressor and comparison of the results with the ones obtained from a different approach (L1 Penalization)
2. Videos' data prediction with different algorithms and export of the results

Thanks to:

stvar: https://stackoverflow.com/a/65899758/25647133

Useful information:

How to find the Channel ID of a YouTube Channel: https://www.youtube.com/watch?v=zNABOcxGkt8