This project leverages the YouTube Data API and Natural Language Processing (NLP) techniques to gain insights into YouTube video recommendations based on user watch history. By analyzing video transcripts and identifying keywords, it sheds light on the algorithm driving video suggestions.
- OAuth 2.0 Integration: Securely authenticate with YouTube API for data access.
- Transcript Analysis: Extract and analyze video transcripts using NLP.
- Keyword Extraction: Identify keywords from transcripts to understand video content.
- Visualization: Visualize insights through charts and word clouds.
- Export Data: Save results in JSON and Excel formats for further analysis.
- Clone the repository.
- Set up OAuth 2.0 credentials on Google Cloud Console.
- Run the data pipeline using Kedro.
- Analyze results, visualize insights, and gain a better understanding of YouTube recommendations.
- Python 3.x
- Kedro
- YouTube Data API
- NLTK
Reporting Dashboard: https://app.powerbi.com/view?r=eyJrIjoiNWQ1NzIwMjEtZDA5MC00NjYxLTlkNTMtNGVkMTgyZjcwMjFkIiwidCI6ImRiNmUxMTgzLTRjNjUtNDA1Yy04MmNlLTdjZDUzZmE2ZTlkYyIsImMiOjEwfQ%3D%3D