🤖 AI-Powered Graduation Project Recommendation System

📌 Overview

This project implements an intelligent AI-powered recommendation and semantic similarity platform for graduation projects using:

• Natural Language Processing (NLP)
• Semantic Search
• Vector Embeddings
• Hybrid Ranking Systems
• Large Language Models (LLMs)

The system helps students:

• discover unique graduation project ideas
• avoid duplicate projects
• analyze originality
• generate intelligent project features
• receive context-aware recommendations through an AI chatbot

---

⚙️ System Pipeline

1️⃣ Data Preprocessing

• Text normalization
• Duplicate removal
• Smart content merging
• Technical keyword extraction
• Feature engineering

2️⃣ Feature Extraction

• KeyBERT-based keyword extraction
• Automatic technical term detection
• Semantic feature generation

3️⃣ Embedding Generation

• SentenceTransformer embeddings
• Normalized vector representations
• Semantic encoding of projects

4️⃣ Semantic Retrieval

• FAISS vector indexing
• Nearest-neighbor semantic search
• Fast project similarity lookup

5️⃣ Hybrid Ranking

The final ranking combines:

• Semantic similarity
• Feature similarity
• Coverage ratio
• Confidence estimation
• Originality analysis

6️⃣ AI Recommendation Engine

• Context-aware project generation
• Feature recommendation
• Novelty checking
• Conversational chatbot assistance

---

🧠 AI & NLP Technologies Used

🔹 Machine Learning & NLP

• SentenceTransformers
• KeyBERT
• Scikit-learn
• SciPy
• FAISS

🔹 LLM Integration

• Google Gemini API
• Ollama
• Mistral

🔹 Backend & Infrastructure

• FastAPI
• Pandas
• NumPy
• Python

---

🏗️ Project Architecture

User Query
    ↓
Intent Classification
    ↓
Context Builder
    ↓
Feature Extraction
    ↓
Embedding Generation
    ↓
FAISS Semantic Search
    ↓
Hybrid Ranking Engine
    ↓
Originality & Duplicate Analysis
    ↓
AI Recommendation Response

---

🔍 Similarity Engine Workflow

Raw Dataset
    ↓
Preprocessing
    ↓
Feature Extraction
    ↓
Sentence Embeddings
    ↓
FAISS Indexing
    ↓
Semantic Retrieval
    ↓
Feature Similarity Matching
    ↓
Hybrid Re-ranking
    ↓
Final Recommendation

---

🚀 Features

✅ AI Chatbot

• Context-aware conversations
• Intent classification
• Domain-specific recommendations
• Memory-aware responses

✅ Semantic Similarity Search

• Embedding-based retrieval
• Semantic duplicate detection
• Vector search with FAISS

✅ Hybrid Recommendation System

• Multi-stage ranking pipeline
• Feature-level semantic comparison
• Adaptive scoring strategy

✅ Originality Detection

• Duplicate risk analysis
• Originality scoring
• Similarity confidence estimation

✅ Intelligent Feature Generation

• AI-generated project features
• Novelty-aware generation
• Domain-aware recommendations

---

📊 Evaluation

The system includes:

• Self-retrieval evaluation
• Real-query testing
• Hybrid ranking validation
• Confidence scoring

Evaluation Metrics

• Semantic Similarity Score
• Hybrid Score
• Originality Score
• Confidence Score
• Duplicate Risk Classification

---

📁 Project Structure

GRADUATION_PROJECT/
│
├── api/                         # FastAPI backend
│
├── Data/
│   ├── raw/                    # Original dataset
│   └── processed/              # Cleaned dataset
│
├── models/                     # FAISS index & metadata
│
├── Notebooks/
│   └── TEST.ipynb              # Training & evaluation notebook
│
├── src/
│   ├── recommendation_engine/  # Chatbot & recommendation logic
│   └── similarity_model/       # Semantic search engine
│
├── requirements.txt
├── README.md
└── .gitignore

---

🧩 Recommendation Engine Modules

recommendation_engine/

Contains:

• Chatbot engine
• Intent classification
• Prompt building
• Idea generation
• Feature generation
• Memory management
• Novelty checking
• Response formatting

---

🔬 Similarity Model Modules

similarity_model/

Contains:

• Semantic search
• Embedding engine
• Hybrid ranker
• Feature similarity engine
• Preprocessing pipeline
• Evaluation framework

---

⚡ Installation

1️⃣ Clone Repository

git clone https://github.com/YOUR_USERNAME/YOUR_REPOSITORY.git
cd YOUR_REPOSITORY

---

2️⃣ Create Virtual Environment

Windows

python -m venv .venv
.venv\Scripts\activate

Linux / Mac

python3 -m venv .venv
source .venv/bin/activate

---

3️⃣ Install Dependencies

pip install -r requirements.txt

---

🔑 Environment Variables

Create a .env file:

GEMINI_API_KEY=your_api_key_here

---

▶️ Running The Project

Run FastAPI Server

uvicorn api.main:app --reload

---

Run Notebook

jupyter notebook

Open:

Notebooks/TEST.ipynb

---

💡 Example Query

Input

AI-based smart library recommendation platform

Output

• Similar graduation projects
• Semantic similarity scores
• Originality analysis
• Duplicate risk estimation
• Recommended features

---

🎯 Future Improvements

• Full RAG integration
• Multi-agent orchestration
• GPU acceleration
• Advanced evaluation metrics
• Real-time deployment
• Database persistence
• Frontend dashboard

---

📚 Research Areas Covered

• Natural Language Processing (NLP)
• Semantic Search
• Recommendation Systems
• Vector Databases
• Conversational AI
• Information Retrieval
• Hybrid Ranking Systems
• Large Language Models (LLMs)

---

👨‍💻 Author

Yossef Assem

---

📄 License

This project is for educational and research purposes.