📚 Book Recommendation System (BRS)

A Book Recommendation System that suggests books to users based on their preferences, genres, and similarity with other titles.
This project applies Natural Language Processing (NLP) and Machine Learning techniques to recommend books using text data such as titles, genres, and descriptions.

---

📖 Table of Contents

• Overview
• Features
• Tech Stack
• Application Pages
• Project Structure
• Setup Instructions
• Running the Application
• Dataset Information
• Model Overview
• Future Improvements
• License
• Author

---

🧠 Overview

The Book Recommendation System (BRS) recommends books based on user interests and genre preferences.
It uses TF-IDF (Term Frequency–Inverse Document Frequency) and Cosine Similarity to analyze book descriptions and find the most similar titles.

The system is implemented with a Streamlit web interface that allows users to:

• Browse books by genre
• Get detailed information about selected books
• View personalized recommendations

---

🚀 Features

✅ Content-based book recommendation using TF-IDF
✅ Genre-based filtering
✅ Dynamic book suggestions and details
✅ Streamlit-based interactive UI
✅ Pre-trained model storage for fast loading
✅ Modular code structure for easy maintenance

---

🧰 Tech Stack

Category	Tools Used
Language	Python 3.x
Framework	Streamlit
Libraries	pandas, numpy, scikit-learn
Dataset	Custom dataset (Books_10000.csv)
Visualization	Streamlit components
Model Files	.pkl files (TF-IDF, similarity matrices, final dataframe)

---

🧩 Application Pages

🏠 1. Home Page

Displays the project introduction, purpose, and navigation to other pages.

---

​🔖 2. Genre Page

Shows the top 50 most famous genres in the data used for suggestions.

---

📜 3. Book Suggestions

Provides 10 random book suggestions according to the genre you choose.

📔 4. Book Details

Displays the book details of the book you choose from the suggestions along with button to get more suggestions from the selected book.

📚 5. Recommendations Page

Show 6 similar recommended books from the book you initially choose from your favourite genre.

---

📁 Project Structure

.
├── .gitignore
├── .streamlit/config.toml # Streamlit app configuration
├── LandingPage.py # Main app entry point
├── downloadModel.py # Downloads model from Google Drive
├── pages/
│ ├── 1_GenrePage.py
│ ├── 2_SuggBooks.py
│ ├── 3_BookDet.py
│ └── 4_Recommendations.py
├── rawData/rawDatasetDownload.py # Script to retrieve raw data
├── recommender_Model/
│ ├── final_df.pkl # Processed book dataset
│ └── tfidf_matrices.pkl # TF-IDF vector data
└── requirements.txt # Dependencies

---

⚙️ Setup Instructions

1️⃣ Clone the Repository

git clone https://github.com/Aaryan10000/Book.Recommendation.System.git
cd Book.Recommendation.System

2️⃣ Create a Virtual Environment

python3 -m venv BRSenv source BRSenv/bin/activate # Linux/macOS

OR

BRSenv\Scripts\activate # Windows

3️⃣ Install Dependencies

pip install -r requirements.txt

4️⃣ Download Model Files

The trained similarity matrices and TF-IDF data are hosted on Google Drive (due to GitHub’s 100MB limit). Run the following command to automatically download them: python downloadModel.py

⚠️ This will place the .pkl model files into the recommender_Model/ folder automatically.

▶️ Running the Application

Start the Streamlit web app:

streamlit run LandingPage.py

This will launch the app in your default browser

🧾 Dataset Information

This project uses the “Best Books Ever” dataset, originally published on Kaggle by mexwell.

📘 License: Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) 📂 Dataset Source: Kaggle - Best Books Ever

⚠️ Usage Note: This dataset is provided strictly for research and educational purposes. You must give appropriate credit to the original author (“mexwell” on Kaggle) and cannot use it for commercial purposes.

🧩 Model Overview

The model uses content-based filtering with the following pipeline:

1. Data Preprocessing → Clean and normalize book descriptions
2. TF-IDF Vectorization → Convert text to numeric representation
3. Cosine Similarity → Compute similarity between books
4. Recommendation Engine → Suggest top 10 similar books All processed matrices are saved as .pkl files in recommender_Model/ for faster access.

💡 Future Improvements

• Add collaborative filtering for user-based recommendations
• Include book cover images using an API (e.g., Google Books API)
• Integrate user login and personalized dashboards
• Deploy using Streamlit Cloud or Render
• Improve NLP model using BERT embeddings

🪪 License

This project is licensed under the MIT License. You are free to use, modify, and distribute this project with attribution.

👨‍💻 Author

Aaryan Dawalkar 💻 B.Tech in Computer Science Engineering 📫 Linkedin