AI-Driven NEET UG Medical Admission Data Entry

# Project Requirement Document

## AI-Based Automated NEET UG Medical Admission Prediction Platform

### Project Overview

We are developing a nationwide AI-powered NEET UG Medical Admission Prediction & Counselling Platform for MBBS admissions across India.

The goal is to create a highly automated system that can:

* Collect counselling data automatically
* Process and standardize data from different states
* Generate accurate college prediction
* Forecast expected closing ranks
* Provide admission probability
* Reduce manual data entry dependency

The system should initially use past 3 years’ counselling data and later automatically support future counselling years with minimal human intervention.

---

# IMPORTANT OBJECTIVE

The system must be designed so that:

✅ After initial setup and training
✅ One admin/operator can manage the entire platform
✅ No large freelancer/data-entry team should be required in future

The platform should automate:

* Data collection
* File downloading
* Data extraction
* Data normalization
* Validation
* Cutoff generation
* Seat matrix processing
* Forecast preparation

---

# Main Platform Goals

## 1. Automated Data Collection

System should automatically collect:

* MCC counselling data
* State counselling data
* Seat matrix
* Round-wise allotment results
* Cutoff files
* Vacancy data

from official counselling websites.

---

# 2. Universal Data Processing System

Different states provide data in different formats:

* Excel
* PDF
* Candidate allotment lists
* Closure reports
* HTML tables

The platform must automatically convert all formats into one universal standardized structure.

---

# 3. AI-Based Prediction System

The system should:

* Analyze previous years’ trends
* Compare seat changes
* Detect cutoff movement
* Forecast expected closing rank
* Calculate admission probability

for each:

* College
* Category
* Quota
* Round

---

# 4. Full College Recommendation Engine

Student inputs:

* AIR
* Category
* State
* Budget
* Domicile
* Preferences

System outputs:

* Safe colleges
* Moderate colleges
* Dream colleges
* Best counselling round strategy
* Expected admission probability

---

# 5. Premium Counselling Dashboard

Dashboard should include:

* Student management
* Prediction reports
* Counselling strategy
* Saved preferences
* Round-wise tracking
* Live cutoff movement
* Vacancy tracking

---

# AUTOMATION REQUIREMENTS

## A. Auto Scraping System (VERY IMPORTANT)

System should automatically:

* Visit counselling websites
* Detect latest files
* Download files
* Store raw files
* Trigger parser automatically

Supported sources:

* MCC
* State counselling authorities
* AIQ
* Deemed universities
* Government medical counselling websites

---

## B. Parser Engine

Separate parser modules should be created for each authority.

Example:

* MCC parser
* Gujarat parser
* Karnataka parser
* Rajasthan parser

The parser should:

* Read PDFs/Excel files
* Extract counselling data
* Identify categories/quota/round
* Generate closing ranks automatically

---

## C. Normalization Engine

The system must standardize:

* Categories
* Quotas
* Rounds
* College names
* Course names

Example:

* GEN → OPEN
* SEBC → OBC
* AIQ → All India Quota

---

## D. Validation Engine

System should automatically detect:

* Duplicate entries
* Invalid ranks
* Missing categories
* Incorrect college mapping
* Wrong scores

Admin should receive validation alerts.

---

# FUTURE-READY REQUIREMENT

The system must NOT be hardcoded only for current years.

It should support:

* Future counselling years
* New colleges
* New quotas
* Reservation changes
* Seat increases
* New states
* Additional courses (BDS/BAMS/BHMS later)

without major redevelopment.

---

# REQUIRED MASTER DATABASE STRUCTURE

The platform should maintain:

* College master database
* Course master database
* Category mapping database
* Quota mapping database
* Seat matrix database
* Cutoff history database
* Vacancy movement database

---

# PREDICTION ENGINE REQUIREMENTS

Prediction engine should:

* Use past 3+ years data
* Analyze historical trends
* Compare seat matrix changes
* Detect competition changes
* Forecast expected closing ranks

Outputs should include:

* Expected closing AIR
* Admission probability %
* Recommended counselling round
* Safe/Moderate/Dream classification

---

# ADMIN PANEL REQUIREMENTS

Admin dashboard should allow:

* File upload
* Parser management
* Error logs
* Validation review
* Manual correction
* College mapping
* Trend analysis
* Prediction override
* Seat matrix management

The system should minimize manual work as much as possible.

---

# SCALABILITY REQUIREMENT

Initial Phase:

* MBBS only

Future Expansion:

* BDS
* BAMS
* BHMS
* BUMS
* Veterinary
* AYUSH counselling

without rebuilding the architecture.

---

# RECOMMENDED TECH STACK

## Backend

* Python FastAPI

## Database

* PostgreSQL

## Scraping

* Playwright + BeautifulSoup

## Data Processing

* Pandas

## Queue System

* Celery + Redis

## Frontend

* Next.js / React

## Cloud Storage

* AWS S3 / DigitalOcean Spaces

---

# IMPORTANT DEVELOPMENT APPROACH

The system should be built as:

* Data-first architecture
* Modular parser system
* Automation-focused platform
* AI-enhanced prediction engine

NOT as a simple static predictor website.

---

# FINAL GOAL

Build a scalable and highly automated NEET UG counselling intelligence platform where:

* One trained operator/admin can manage the system
* Manual data entry dependency is minimal
* Future counselling years can be processed automatically
* Prediction quality continuously improves with data
* Platform becomes a premium counselling and admission solution for students across India