Final Project Proposal

04 Nov 2025

---

Mānoa Hunting — A personalized guide to good restaurants around UH

Overview

The problem

UH Mānoa students, staff, and neighbors are surrounded by great food—but discovering the right spot at the right time is hard. Instagram is noisy, Google Maps is generic, and group chats get outdated. We need a way to filter by budget, dietary needs, distance/wait time, parking/bus access, and even campus schedule (class blocks, club meetings, labs).

The solution

Mānoa Hunting is a Next.js + React + Bootstrap 5 web app that builds a personal taste profile for each logged-in user and recommends nearby restaurants and food trucks around UH Mānoa (and greater Oʻahu) with live crowd-sourced “vibes” (busy/quiet), today’s specials, student deals, and time-to-eat suggestions between classes.

Our “special sauce”: after registration, users answer a short taste quiz (price comfort zone, diet, spice tolerance, allergens, cuisines, caffeine needs, time windows). The app learns from their bookmarks, ratings, and “nope” list to deliver hyper-personalized picks and smart notifications (e.g., “You’ve got 55 minutes before ICS 314 — two top picks within a 6-minute walk, average wait 8 minutes.”).

Tech & hosting: Next.js (App Router), React, Bootstrap 5 UI, MongoDB (profiles, places, reviews), NextAuth (email or UH Google SSO), Map/Geolocation APIs for distance & routing, CI/CD from GitHub (public repo; docs via GitHub Pages; deployment from GitHub to Vercel).

---

Names of the proposers

• Shuwei Chen

---

Mockup page ideas

1. Landing Page
   • Hero: “Find your next bite near UH.”
   • Quick filters: Budget ($/\(/\)$), Distance (walk/BIKESHARE/bus), Diet (vegetarian, vegan, halal, gluten-free), Vibe (quiet to study, lively with friends).
   • “Sign in to personalize” CTA.
2. Discover (Personalized Feed)
   • Card grid of nearby restaurants ranked by your taste profile.
   • Chips showing fit reasons (e.g., “close to POST,” “spicy level 2,” “under $12,” “open now”).
   • Live “vibe meter” (crowd-sourced check-ins + historical patterns).
3. Map View
   • Map with pins, walking times from your current building (e.g., POST, Hamilton, Campus Center).
   • Toggle: “Between classes” (shows only places that fit your available window).
4. Place Details
   • Menu highlights, dietary flags, student deals, today’s specials, average wait, parking & bus notes (routes 4/13/6…), user photos.
   • Buttons: “Add to Favorites,” “I’m Here,” “Queue vibes: Busy/Chill.”
5. My Profile
   • Taste quiz (cuisines, spice, allergens, price), saved places, hidden places, notification windows (e.g., weekdays 11:00–14:00).
   • “Taste Twin” section: people with similar palates (anonymized) and what they loved this week.
6. Admin Dashboard
   • Verify businesses, moderate photos/reviews, manage specials & tags, merge duplicates.

---

Use case ideas

• New user → first lunch
  1. User hits Landing, signs in.
  2. Completes 1-minute taste quiz; grants geolocation.
  3. Discover shows 8 nearby cards tailored to budget + diet; “Between classes: 55 minutes” filter on by default.
  4. User opens Place Details → sees “average wait 7–10 min,” “$11 lunch special,” “5-minute walk from POST.”
  5. User bookmarks, walks using Map View, taps “I’m Here,” and later rates + flags favorites.
• Commuter with dietary restriction
  1. Profile set to gluten-free and halal preferred.
  2. Discover prioritizes compliant menus and shows “trust score” based on community tags/photos.
  3. User subscribes to notifications 10:30–12:30 on Tue/Thu; receives ping when a compliant place near Campus Center is calm.

---

Beyond the basics

• Crowd-sourced Vibe & Wait Time Lightweight check-ins (“Chill / Normal / Busy”), auto-expires. Historical smoothing predicts lunch rush near class transition times.

• Smart Time Windows Parse your optional class schedule blocks (or manual time windows) to suggest places that fit your precise gaps, including walk time + queue + eat time.

• Transit & Parking Hints Quick view: bus lines, bike racks, meter/lot notes, and known “gotchas” (cash-only windows, last call times).

• Student Deals & Loyalty Owners can post UH-specific specials; users can track stamp/punch progress (QR or one-tap “redeemed”), with guardrails to prevent abuse.

• Evaluation plan (local community) Recruit UH students (ICS 314 classmates, clubs) and nearby vendors on University Ave/King St to test recommendations, vibe accuracy, and deal redemption. Gather usability metrics and iterate.

---

Approach

• Stack: Next.js (App Router), React, Bootstrap 5 components, MongoDB Atlas, NextAuth, Map & Geolocation APIs.
• Data model: User, TasteProfile, Place, MenuItem, Special, CheckIn, Review, Favorite.
• Special sauce state: Per-user taste vectors, favorites/hidden, notification windows, class-gap windows, and “Taste Twin” cohorts.
• DevOps: Public GitHub repo with Issues/Projects, ESLint/Prettier, GitHub Actions CI, preview deploys; app deployed from GitHub to Vercel; portfolio essay published via GitHub Pages (TechFolio).
• Testing: Playwright for E2E (critical paths: sign-in, discovery, place details, check-ins), Jest/RTL for components, seed data for local dev.

---