Book Recommender System using Classical Machine Learning Techniques

In this project, we have developed a book recommender system using a combination of five different techniques to enhance recommendation accuracy and provide diverse recommendation strategies. The techniques used are: Techniques Used

1. Decision Trees (DT)
2. K-Nearest Neighbors (KNN)
3. Random Forest (RF)
4. LightGBM
5. Collaborative Filtering

1. Decision Trees (DT)

Decision Trees are used to model user preferences by learning decision rules from the data. They are easy to interpret and provide a clear visualization of the decision-making process.

2. K-Nearest Neighbors (KNN)

KNN is a non-parametric method that predicts the rating of a book based on the ratings of its nearest neighbors. It is simple to implement and effective for small datasets.

3. Random Forest (RF)

Random Forest is an ensemble learning method that combines multiple decision trees to improve prediction accuracy and control overfitting. It provides robust performance for various types of data.

4. LightGBM

LightGBM is a gradient boosting framework that uses tree-based learning algorithms. It is designed to be distributed and efficient with faster training speed and higher efficiency compared to traditional boosting algorithms.

5. Collaborative Filtering

Collaborative Filtering uses the preferences of similar users to make recommendations. It is particularly effective for recommendation systems as it captures the underlying patterns in user behavior.

Conclusion

The F1 score comparison highlights that Collaborative Filtering is the best technique for this book recommender system. Its ability to utilize user similarity and past preferences results in highly accurate recommendations. While other techniques like LightGBM and Random Forest also performed well, Collaborative Filtering's specialized approach for recommendation tasks gave it an edge, making it the preferred choice for this project.