Note

## To optimize:

1. Use Pre-filtering

   - Instead of comparing every lost item with every found item, apply filters:
   - Same category (e.g., only match backpacks with backpacks).
   - Same timeframe (limit searches to a few weeks).
   - Nearby locations (reduce unnecessary comparisons).

2. Optimize Matching Frequency

   - Run real-time matching only for new submissions.
   - Re-run matches periodically for older items instead of every time a new item is added.

3. Use a Scoring Threshold
   - Only store matches with a high AI confidence score (e.g., 80%+).
   - Avoid creating too many unnecessary Match entries.

## Final Flow (Efficient Matching Process)

- User A submits a lost item.
- User B submits a found item (it’s just an independent post).
- AI matches lost & found items only when necessary based on pre-filters.
- If a high-confidence match is found → create a Match & notify users.

Tickets:
Upload Lost Item
Upload Found Item
AI-Powered Tagging
AI Matching - Pre-filtering
AI Matching - Batch Processing
AI Matching - Scoring System
Match Notifications
User Authentication
Cloud Storage
AI Optimization
Periodic Matching Updates
UI/UX Design