This project focuses on building a Book Recommendation System using collaborative filtering techniques. Collaborative filtering is an unsupervised learning approach for recommendation tasks, aiming to predict user preferences by identifying patterns or similarities in user behavior or item features.
The system employs the k-nearest neighbors (KNN) algorithm, which is effective for finding similar books based on user preferences or book characteristics. This instance-based learning method is commonly used in recommendation systems for its ability to identify related items.
The dataset, obtained from Kaggle, contains 12 columns and 11,126 unique entries. You can find it here.
- Data exploration and visualization
- Model built with KNN algorithm
- Example output includes book recommendations with covers
Clone the repository and run the notebook to generate book recommendations based on user or book attributes.