How chaitanya697/noteseek works
NoteSeek is a Flask-backed search and study-notes analysis application. A user submits queries or requests other analysis through the browser; Flask routes them to the relevant engine and returns results to the interface. The engine covers several distinct retrieval models, classification, clustering, recommendations, evaluation, and document ingestion into a live index. The index and preprocessing layer underpin multiple capabilities. Relationships among unsampled components are included only where documented by the README or named in the source index; their internal call details were not inspected.
Main entry points: app.py Flask routes · indexer.py Inverted index · preprocessor.py NLP preprocessing · uploader.py Document extraction
Components 16
Web application
- Browser interface
static/js/main.js - Flask routes
app.py
Search and indexing
- NLP preprocessing
engine/preprocessor.py - Inverted index
engine/indexer.py - VSM and BM25
engine/retriever.py - Boolean search
engine/boolean_model.py - Language models
engine/language_model.py
Document analysis
- Classification
engine/classifier.py - Clustering
engine/clusterer.py - Evaluation metrics
engine/evaluator.py
Ingestion and recommendations
- Document extraction
engine/uploader.py - Recommendations
engine/recommender.py
Notes and state
- Notes collection
data/notes.json - Uploaded files
uploads/.gitkeep
Other components
- Student
- PDF/PPTX files
Connections 30
- Student → Browser interface: submits requests
README.md - Browser interface → Flask routes: sends requests
README.md - Flask routes → VSM and BM25: runs search
app.py - Flask routes → Boolean search: runs search
app.py - Flask routes → Language models: routes search
README.md - Flask routes → Classification: classifies text
app.py - Flask routes → Clustering: runs clustering
app.py - Flask routes → Evaluation metrics: routes evaluation
README.md - Flask routes → Recommendations: gets suggestions
app.py - Flask routes → Document extraction: processes uploads
app.py - Flask routes → Inverted index: serves documents
README.md - Notes collection → Inverted index: provides notes
README.md - Inverted index → NLP preprocessing: preprocesses text
engine/indexer.py - VSM and BM25 → NLP preprocessing: preprocesses queries
engine/retriever.py - Boolean search → NLP preprocessing: processes terms
engine/boolean_model.py - Language models → NLP preprocessing: processes queries
engine/language_model.py - Classification → NLP preprocessing: preprocesses text
engine/classifier.py - Clustering → NLP preprocessing: preprocesses text
engine/clusterer.py - Recommendations → NLP preprocessing: preprocesses queries
engine/recommender.py - Document extraction → NLP preprocessing: preprocesses text
engine/uploader.py - VSM and BM25 → Inverted index: reads index
engine/retriever.py - Boolean search → Inverted index: reads postings
engine/boolean_model.py - Language models → Inverted index: reads term counts
engine/language_model.py - Classification → Inverted index: uses document vectors
engine/classifier.py - Clustering → Inverted index: uses document vectors
engine/clusterer.py - Recommendations → Inverted index: uses document vectors
engine/recommender.py - PDF/PPTX files → Document extraction: provides files
README.md - Document extraction → Uploaded files: saves uploads
- Document extraction → Inverted index: adds documents
engine/uploader.py - VSM and BM25 → Recommendations: logs interactions
app.py