How m-rishab/research-paper-recommendation works
This is a research-paper recommendation prototype. Its documented workflow takes a paper title, uses NLP and MiniLM to predict a subject area, and returns five recommended papers; the README also describes preprocessing and model training. The tree contains a notebook, saved vectorizer artifacts, and a web template, but no app.py or source excerpts. Thus the runtime wiring among them cannot be verified; the links below reflect documented capabilities, not inspected calls.
Main entry points: RPR.ipynb Model workflow · index.html Title form · Models Vectorizer artifacts
Components 6
User interface
- Title form
templates/index.html
Recommendation and NLP
- Model workflow
Notebook/RPR.ipynb - Top-five recommendations
Notebook/RPR.ipynb - Subject prediction
Notebook/RPR.ipynb
Saved model artifacts
- Vectorizer artifacts
Notebook/Models
Other components
- Paper seeker
Connections 5
- Paper seeker → Title form: enters title
README.md - Title form → Model workflow: submits title
- Model workflow → Top-five recommendations: produces top five
README.md - Model workflow → Subject prediction: predicts area
README.md - Model workflow → Vectorizer artifacts: uses artifacts