-
-
Home page layout with 3d model car for better ui.
-
Dashboard part 1 where user can input text which will be processed and will properly inserted in the database.
-
Dashboard part 2 where user input a specific company name or model and gets insight of that, it contains ai generated summary.
-
Dashboard part 3 where we see whether the company improving or not also it show positive review last 30 days and the change made by company.
-
Dashboard part 4 where we compare models and get the best based on review.
-
Dashboard part 5 where we show from where our user give the review on the map, also shows the recent review on which company at the top.
-
Dashboard part 6 shows radical graph for specific company and bar chart of all the reviews of all company.
-
Competitive intelligence which helps company to do deep analysis on there and other models.
-
Last is our team info.
🚗 GeoDrive Intelligence
📌 Inspiration
The idea behind GeoDrive Intelligence was driven by a real problem: automotive companies are drowning in scattered, unstructured customer feedback.
Reviews exist across multiple platforms, making it difficult to extract meaningful insights quickly. Most companies rely on manual analysis or basic tools that fail to capture real-time trends.
We built this to convert raw, messy data into clear, decision-ready intelligence.
⚙️ What it does
GeoDrive Intelligence is an AI-powered platform that transforms automotive reviews into actionable insights.
Key Features:
- Performs sentiment analysis on customer reviews
- Extracts key topics like:
- Mileage
- Engine performance
- Safety
- Mileage
- Provides:
- 📊 Interactive dashboards
- 🌍 Geospatial insights using maps
- 📊 Interactive dashboards
- Enables:
- Brand comparison
- Market trend analysis
- Brand comparison
🛠️ How we built it
Frontend
- Built using React (Vite) for speed and responsiveness
Backend
- Developed with FastAPI for efficient API handling
- Integrated AI models for sentiment analysis
Visualization Tools
- Chart.js → Analytics dashboards
- Leaflet.js → Geospatial mapping
Architecture
Below is the high-level system architecture showing the flow from user input to AI-driven insights and visualization:

Deployment
- Hosted on Vercel for seamless access
⚠️ Challenges we ran into
- Deployment issues (white screen errors)
- API connection problems (localhost vs deployed backend)
- CORS issues between frontend and backend
- SPA routing issues on Vercel
- Git merge conflicts during team collaboration
🏆 Accomplishments that we're proud of
- Built a fully functional full-stack AI application
- Implemented real-time sentiment analysis
- Integrated geospatial intelligence
- Successfully deployed both frontend and backend
- Solved real-world deployment and integration issues
📚 What we learned
- Practical implementation of AI-based sentiment analysis
- Full-stack debugging in production environments
- Deployment workflows using Vercel
- Handling real-world issues like:
- CORS
- Routing
- CORS
- Effective team collaboration using Git
🚀 What's next for GeoDrive Intelligence
- Integrate live data sources (social media, review platforms)
- Improve AI models for deeper insights
- Add predictive analytics for market trends
Example (future predictive model):
$$ Trend = \alpha \cdot Sentiment + \beta \cdot Volume + \gamma \cdot Time $$
- Enhance UI/UX for better usability
- Expand beyond automotive into multiple industries
✅ Task Roadmap
- Build core platform
- Implement sentiment analysis
- Deploy application
- Add live data integration
- Improve prediction models
- Expand to other industries
Built With
- analysis
- chart.js
- css
- fastapi
- git
- github
- html
- javascript
- leaflet.js
- python
- react-(vite)
- rest-api
- sentiment
- vercel
Log in or sign up for Devpost to join the conversation.