Book Recommendation System

A Content-Based Recommendation System built using Python that suggests books to users based on their description. The system utilizes a Bag of Words (BoW) model and cosine similarity to find books similar to the one selected by the user. The project also includes a Streamlit-based web application to provide an interactive interface for recommendations.

Features

1. Content-Based Filtering: Recommends books based on textual descriptions.
2. Bag of Words Model: Represents the text data.
3. Cosine Similarity: Measures the similarity between book descriptions.
4. Streamlit Application: Provides an easy-to-use web interface.

---

Methodology

1. Data Preparation

• Collect a dataset containing books and their descriptions.
• Preprocess the text data by:
  • Removing stop words.
  • Tokenizing the text.
  • Stemming/lemmatizing.

2. Bag of Words Model

The Bag of Words (BoW) model is created to represent the textual data. This involves:
BoW Vector = [Frequency of Term 1, Frequency of Term 2, ..., Frequency of Term N]
Where N is the size of the vocabulary.

Example Representation:

"The Hobbit": [1, 2, 0, ..., 1]
"The Lord of the Rings": [0, 1, 3, ..., 2]

Each vector corresponds to the frequency of terms in the text.

3. Cosine Similarity

To measure the similarity between book descriptions, we calculate cosine similarity:
Cosine Similarity = (A · B) / (||A|| ||B||)

Where:

• A and B are the vectors of two book descriptions.
• ||A|| and ||B|| are the magnitudes (norms) of the vectors.

Example Calculation:

A = [1, 2, 1], B = [2, 1, 3]
Dot Product (A . B) = (1*2) + (2*1) + (1*3) = 7
Magnitude of A (||A||) = sqrt(1^2 + 2^2 + 1^2) = sqrt(6)
Magnitude of B (||B||) = sqrt(2^2 + 1^2 + 3^2) = sqrt(14)
Cosine Similarity = 7 / (sqrt(6) * sqrt(14)) ≈ 0.89

4. Recommendation Process

• Compute the cosine similarity between the user's selected book and all other books.
• Rank the books based on similarity scores.
• Display the top k recommendations to the user.

---

Web Application (Streamlit)

Home Page

• Displays the list of available books for selection.
• Allows users to choose a book to get recommendations.

Recommendation Page

• Shows the top k books similar to the selected book.
• Displays similarity scores alongside recommendations.

---

---

Installation

Prerequisites

• Python 3.10 or above
• Required Python libraries:
  pip install streamlit pandas scikit-learn numpy

Running the Application

1. Clone the repository:

   git clone https://github.com/Srujanrana07/book-recomender-system.git
   cd book-recommendation-system

2. Run the Streamlit app:

   streamlit run app.py

---

Example Workflow

1. User selects a book (e.g., "The Hobbit").
2. System processes the description and computes cosine similarity.
3. Recommended books based on similarity are displayed:
   • "The Lord of the Rings"
   • "Harry Potter and the Sorcerer's Stone"
   • "Percy Jackson & The Olympians"

---

Mathematical Insights

Text Representation (BoW):

Vector for "The Hobbit" = [1, 2, 0, ..., 1]
Where each element represents the frequency of a specific word.

Cosine Similarity Calculation Example:

A = [1, 2, 1], B = [2, 1, 3]
Cosine Similarity = ((12) + (21) + (1*3)) / sqrt(1^2 + 2^2 + 1^2) * sqrt(2^2 + 1^2 + 3^2) = 0.89

---

Future Improvements

1. Incorporate user feedback for better personalization.
2. Extend to hybrid recommendation systems.
3. Use advanced NLP techniques like TF-IDF or word embeddings for better text representation.

---

License

This project is licensed under the MIT License.

---

Live Demo