How prasanna-ks-85/youtube-intelligence-platform works
This is a YouTube video analytics backend: a user submits a video URL, the application invokes an analysis pipeline, and the pipeline returns structured analysis and visualizations. The pipeline’s documented stages cover extraction, preprocessing, classification, summarization, analytics, keyphrases, insights, topics, emotions, explainability, and executive intelligence. Settings, caching, and shared schemas form supporting layers. Some pipeline calls and the application’s interface are not shown in the excerpts; those relationships are marked unverified rather than asserted as observed calls.
Main entry points: modules/pipeline.py Pipeline orchestrator · extractor.py YouTube extraction · schemas/pipeline.py Pipeline result schema · app.py Application
Components 21
Application entry
- Application
app.py
Analysis workflow
- Pipeline orchestrator
modules/pipeline.py - YouTube extraction
modules/extractor.py - Text preprocessing
modules/preprocessor.py - Keyphrase extraction
modules/keyphrase_extractor.py - Video classification
modules/classifier.py - Video summarization
modules/summarizer.py - Audience analytics
modules/analytics.py - Executive insights
modules/insights.py - Topic discovery
modules/topic_modeler.py - Emotion analysis
modules/emotion_analyzer.py - Classification explanation
modules/explainability.py - Executive intelligence
modules/executive_intelligence.py - Visualizations
modules/visualizer.py
Data and support
- Analysis cache
utils/cache.py - Runtime settings
config/settings.py
Result contracts
- Core data schemas
schemas/models.py - Pipeline result schema
schemas/pipeline.py - Analysis result schemas
schemas
Other components
- Analyst
- YouTube services
Connections 19
- Analyst → Application: submits URL
- Application → Pipeline orchestrator: invokes analysis
app.py - Pipeline orchestrator → YouTube extraction: extracts video
modules/pipeline.py - YouTube extraction → YouTube services: fetches video data
modules/extractor.py - YouTube extraction → Analysis cache: reads and writes
modules/extractor.py - Pipeline orchestrator → Text preprocessing: processes text
modules/pipeline.py - Pipeline orchestrator → Keyphrase extraction: extracts phrases
modules/pipeline.py - Pipeline orchestrator → Video classification: classifies video
modules/pipeline.py - Pipeline orchestrator → Video summarization: summarizes video
modules/pipeline.py - Pipeline orchestrator → Audience analytics: computes analytics
modules/pipeline.py - Pipeline orchestrator → Executive insights: generates insights
modules/pipeline.py - Pipeline orchestrator → Topic discovery: discovers topics
modules/pipeline.py - Pipeline orchestrator → Emotion analysis: analyzes emotions
modules/pipeline.py - Pipeline orchestrator → Classification explanation: explains classification
modules/pipeline.py - Pipeline orchestrator → Executive intelligence: generates intelligence
modules/pipeline.py - Pipeline orchestrator → Visualizations: generates charts
modules/pipeline.py - Pipeline orchestrator → Pipeline result schema: returns result
modules/pipeline.py - Pipeline result schema → Core data schemas: uses core types
schemas/pipeline.py - Pipeline result schema → Analysis result schemas: uses analysis types
schemas/pipeline.py