Mess Menu Feedback and Rating System

Mess food is the most complained-about and least measured service on any residential campus. Committees argue from anecdotes — "the dal is always bad" — while contractors dismiss complaints as a vocal minority, because nobody has numbers. This project builds a mess feedback system: the weekly menu published in advance with historical dish ratings, per-meal star ratings from verified residents with structured tags (tasty, bland, served late, less quantity) and comments, analytics by meal slot, day and dish, automatic flagging of dishes below threshold for two consecutive weeks, and a monthly contractor scorecard. Committee responses are published against feedback, so students see that rating causes change.

1. The mess committee enters the weekly menu; residents browse it with each dish's historical rating.
2. After each meal window opens, verified residents submit a star rating with tags and optional comments — one per meal per student.
3. The system aggregates ratings into rolling averages by dish, slot and day, updating dashboards live.
4. Dishes scoring below threshold for two consecutive weeks are auto-flagged for the committee's menu review.
5. The committee publishes responses to notable feedback, visible to all residents.
6. At month end, the contractor scorecard is generated: ratings, trends, flagged dishes and response log.

Tech stack:

• Python (Django/Flask) or Node.js backend
• PostgreSQL/MySQL relational database
• HTML5, CSS3, JavaScript (mobile-first frontend)
• Scheduled jobs for meal-window open/close
• Charting library for analytics
• Role-based access (resident, committee, contractor view)
• CSV/PDF scorecard export
• Responsive web UI

• [Weekly menu publishing] Breakfast, lunch, snacks and dinner for each day, with each dish carrying its historical rating.
• [One-rating-per-meal feedback] Verified residents rate each meal once — fairness is enforced at the data layer, not by policy.
• [Structured tags + comments] Quick tags (tasty, fresh rotis, bland, served late, less quantity) plus free-text comments for the committee.
• [Dish-level analytics] Rolling averages by dish, meal slot and day, with trend views the committee can actually read.
• [Auto-flagging] Dishes below the rating threshold for two consecutive weeks are flagged for menu review automatically.
• [Contractor scorecard] Monthly export: overall rating, slot-wise scores, complaint counts and resolution status.
• [Committee responses] Official replies published against feedback threads, visible to all residents.

• Complete mess feedback application source code
• Database schema and migration scripts
• Menu management and rating-window scheduler
• Analytics dashboards and auto-flagging engine
• Contractor scorecard generator
• Setup and deployment guide
• Pilot plan for the buyer to measure real adoption in one hostel
• Project report PDF (background, feedback design, analytics model, methodology)
• PPT presentation for final review
• Viva Q&A preparation document (rating integrity, aggregation, scheduling)

• Ratings reflect the opinions of residents who choose to rate — participation bias is real and documented in the report.
• One-rating-per-meal is enforced per account; account sharing is a policy matter.
• The system measures satisfaction; it does not inspect kitchens or verify hygiene.
• Menu decisions stay with the committee — analytics advise, they do not auto-change menus.
• Scorecards are only as honest as the rating base; pilot adoption targets are design goals.
• Nutrition tracking and pre-order counts are future scope.