# WayBack

Context-aware re-finding system for personal tourism information.

TUM Lab Course SS 2026 · Project [W4]

## Concept

WayBack surfaces tourism items (saved places, bookmarks, tickets, notes) the user has already seen, at the moment they become relevant again — based on current location and time. Unlike conventional recommenders that suggest new items, this is a personal information re-finding system.

Based on Sappelli, Verberne & Kraaij (2017): Evaluation of context-aware recommendation systems for information re-finding.

## Repository structure

- backend/ Flask + SQLAlchemy + SQLite. Three recommendation methods (CBR, JITIR, CIA).

- frontend/ React + Vite. Mobile-first web app.

- docs/ Specs, API contract, paper notes.

- mocks/ Sample JSON payloads for frontend development.

## Quick start

Backend:

cd backend

python -m venv .venv

.venv\Scripts\activate

pip install -r requirements.txt

python seed.py

python app.py

Frontend:

cd frontend

npm install

npm run dev

## Recommendation methods

- CBR Content-based with TF-IDF + cosine similarity. Strength: context relevance.

- JITIR Just-in-time IR — context as search query. Strength: document relevance.

- CIA Contextual Interactive Activation (3-layer spreading activation). Strength: action prediction + diversity.

## Reference

Sappelli, M., Verberne, S., & Kraaij, W. (2017). Evaluation of context-aware recommendation systems for information re-finding. Journal of the Association for Information Science and Technology, 68(4), 895–910.