https://github.com/alkaifaftab000/clickbait
Automating YouTube Clickbait Detection using sentiment analysis, metadata, and thumbnails to classify video content effectively
https://github.com/alkaifaftab000/clickbait
clickbait clickbait-detection fake-video-detection nltk sentiment-analysis text-blob vader youtube-api
Last synced: 3 months ago
JSON representation
Automating YouTube Clickbait Detection using sentiment analysis, metadata, and thumbnails to classify video content effectively
- Host: GitHub
- URL: https://github.com/alkaifaftab000/clickbait
- Owner: alkaifaftab000
- Created: 2025-03-13T01:23:38.000Z (over 1 year ago)
- Default Branch: master
- Last Pushed: 2025-03-16T11:30:45.000Z (about 1 year ago)
- Last Synced: 2025-06-11T03:06:34.970Z (about 1 year ago)
- Topics: clickbait, clickbait-detection, fake-video-detection, nltk, sentiment-analysis, text-blob, vader, youtube-api
- Language: Jupyter Notebook
- Homepage:
- Size: 37.4 MB
- Stars: 0
- Watchers: 1
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# YouTube Clickbait Detection 🎥🧐
## Project Overview 📜
This project focuses on detecting clickbait content in YouTube videos by analyzing metadata, sentiment, and thumbnails. By utilizing natural language processing and pretrained transformers, it aims to classify videos as clickbait or non-clickbait effectively.
---
## Workflow ⚙️
1. **Fetching data using YouTube API**:
- Retrieved metadata from **750 videos**, including comments, views, likes, thumbnails, and published dates.
2. **Analyzing sentiments**:
- Used tools like **Vader**, **TextBlob**, and **NLTK** for sentiment analysis.
- Implemented **majority voting** to determine the overall sentiment for each video.
3. **Clickbait evaluation**:
- Combined sentiment analysis results and thumbnail information to declare whether a video is clickbait or not.
---
## Progress 🚀
- Identifying sentiments using pretrained transformers like **DeepSeek** and **Llama** to enhance the analysis.
---
## Learning Outcomes 🎓
1. Enhanced skills in using APIs for data extraction and handling large datasets effectively.
2. Gained practical experience in leveraging NLP techniques and transformers for sentiment analysis and classification.
---
## Plugins and Libraries Used 🛠️
- `ollama`
- `youtube-developer`
- `nltk`
- `vader`
- `TextBlob`