A Full-Stack B2B SaaS Platform for Restaurant Review Intelligence
TableTalk is a comprehensive B2B SaaS solution that transforms scattered customer reviews into actionable business intelligence for restaurant owners. By leveraging AI-powered analysis, automated web scraping, and intelligent dashboards, TableTrack helps restaurants understand customer feedback, identify improvement opportunities, and make data-driven decisions to enhance their business performance.
Restaurant owners highly value customer feedback, especially from Google Reviews, but this feedback is often:
- Scattered across multiple platforms
- Difficult to analyze at scale
- Time-consuming to extract actionable insights
- Hard to track trends and performance metrics
TableTalk solves these challenges by providing:
- Automated review aggregation from Google Maps and other platforms
- AI-powered sentiment analysis and entity extraction using Google Cloud NLP
- Real-time performance dashboards with comprehensive analytics
- Actionable AI insights generated using advanced language models
- Smart Review Classification: Categorize reviews by sentiment and topic
- AI-Powered Sentiment Analysis: Track customer satisfaction with Google Cloud NLP
- Multi-Platform Aggregation: Scrape Google Maps using Selenium
- Entity Extraction: Identify key topics, menu items, and business aspects
- Analytics Dashboard: Centralized view of all reviews
- Performance Metrics Tracking: Monitor rating trends, review volume, and sentiment over time
- Review Segmentation: Segment reviews into Highly Positive, Critical, and Suggestions
- AI-Generated Insights: Business recommendations powered by DeepSeek LLM
- Topic-Based Ratings: Track performance across specific aspects of the business
- Critical Review Detection: Highlight and prioritize negative feedback
┌─────────────────────┐ JWT Auth ┌─────────────────────┐ API Calls ┌─────────────────────┐
│ Frontend │ + REST API │ Backend API │ │ External APIs │
│ React.js + │◄───────────────►│ Flask + │◄───────────────►│ │
│ Tailwind CSS │ │ SQLAlchemy │ │ ┌───────────────┐ │
│ │ │ │ │ │ Google Cloud │ │
│ ┌─────────────────┐ │ │ ┌─────────────────┐ │ │ │ NLP API │ │
│ │ Dashboard UI │ │ │ │ Auth Service │ │ │ │ • Sentiment │ │
│ │ Analytics │ │ │ │ Business Logic │ │ │ │ • Entities │ │
│ │ Review Mgmt │ │ │ │ Dashboard APIs │ │ │ └───────────────┘ │
│ └─────────────────┘ │ │ └─────────────────┘ │ │ │
└─────────────────────┘ └─────────────────────┘ │ ┌───────────────┐ │
│ │ │ OpenRouter │ │
│ │ │ (DeepSeek V3) │ │
│ │ │ • AI Insights │ │
│ │ │ • Summary │ │
▼ │ └───────────────┘ │
┌─────────────────────┐ └─────────────────────┘
│ Data Processing │ │
│ Pipeline │ │
└─────────────────────┘ │
│ │
▼ │
┌─────────────────────┐ Web Scraping ┌─────────────────────┐ Raw Data ┌─────────────────────┐
│ Data Sources │◄────────────────│ Scraping Layer │───────────────►│ Database Layer │
│ │ │ │ │ │
│ ┌─────────────────┐ │ │ ┌─────────────────┐ │ │ ┌─────────────────┐ │
│ │ Google Maps │ │ │ │ Selenium │ │ │ │ PostgreSQL/ │ │
│ │ Review Pages │ │ │ │ WebDriver │ │ │ │ MySQL │ │
│ │ │ │ │ │ • Rate Limiting │ │ │ │ │ │
│ └─────────────────┘ │ │ │ • Smart Parsing │ │ │ │ ┌─────────────┐ │ │
│ │ │ └─────────────────┘ │ │ │ │ Tables │ │ │
│ ┌─────────────────┐ │ │ │ │ │ │ • Users │ │ │
│ │ Other Review │ │ │ ┌─────────────────┐ │ │ │ │ • Business │ │ │
│ │ Platforms │ │ │ │ BeautifulSoup │ │ │ │ │ • Reviews │ │ │
│ │ (Future) │ │ │ │ • HTML Parsing │ │ │ │ │ • Insights │ │ │
│ └─────────────────┘ │ │ │ • Data Cleaning │ │ │ │ └─────────────┘ │ │
└─────────────────────┘ │ └─────────────────┘ │ │ └─────────────────┘ │
└─────────────────────┘ └─────────────────────┘
| Layer | Technology |
|---|---|
| Frontend | React.js, Tailwind CSS |
| Backend | Python Flask, SQLAlchemy |
| Database | PostgreSQL/MySQL |
| AI & NLP | Google Cloud Natural Language API, DeepSeek V3 |
| Web Scraping | Selenium WebDriver, BeautifulSoup |
| Authentication | Flask-JWT-Extended |
| API Integration | OpenRouter API |
| Deployment | TBD (AWS/Google Cloud Platform) |
- Phase 1: Research & Market Analysis (✅ Completed)
- Phase 2: Core Backend Development & API Integration (✅ Completed)
- Phase 3: AI/ML Model Development & Training (✅ Completed)
- Phase 4: Frontend Dashboard Development (🔄 In Progress)
- Phase 5: Beta Testing & User Feedback (🔜 Upcoming)
- Phase 6: Production Deployment & Launch (🔜 Upcoming)
- Automated review scraping from Google Maps using Selenium
- Real-time sentiment analysis with Google Cloud NLP
- AI-powered insights generation using DeepSeek LLM
- Comprehensive dashboard with 8+ analytics endpoints
- Primary: Independent restaurant owners (1-10 locations)
- Secondary: Restaurant chains and hospitality businesses
- Tertiary: Food service management companies
- JWT-based authentication system
- Secure password hashing with Werkzeug
- Environment-based configuration management
- Input validation and SQL injection prevention
| Name | Role | GitHub |
|---|---|---|
| Ahmed Abdulla | Team Lead | @ahmed-52 |
| Farhan Mashrur | Team Lead | @farhan-439 |
| Arsh Singh | Developer | @Arsh-S |
| Saesha Agarwal | Developer | @saeshaagarwal |
| Sonja Wong | Developer | @wsonja |
| Daniel Khanin | Developer | @Danikk-392 |
| Arianna Hsu | Developer |
All Rights Reserved - This code is proprietary and confidential. Unauthorized copying, modification, distribution, or use of this software is strictly prohibited without explicit written permission from the authors.
Built as a project by Cornell Data Strategy