This project implements a sophisticated recommendation system leveraging user-based and item-based collaborative filtering techniques. Designed in JavaScript with Node.js, it offers a scalable and efficient approach to predict user preferences and item ratings, based on historical data. The system uses Pearson Correlation Coefficient and Cosine Similarity metrics to calculate similarities and predict ratings, ensuring high accuracy and reliability.
- User-Based Collaborative Filtering: Calculates recommendations based on user similarity and past user ratings.
- Item-Based Collaborative Filtering: Generates recommendations based on item similarity and user-item interaction history.
- Similarity Metrics: Utilizes Pearson Correlation and Cosine Similarity for accurate similarity measurements.
- Dynamic Neighborhood Size: Adjusts the neighborhood size for optimal prediction accuracy.
- Mean Absolute Error (MAE) Calculation: Evaluates the accuracy of predictions.
- JavaScript
- Node.js
To get started with this project, clone the repository and install the necessary dependencies.
git clone https://github.com/abhai28/Reccomendation-System.git
cd Reccomendation-System
Run the system with the following command:
node App.js
Contributions to enhance the functionality or efficiency of this recommendation system are welcome. Please feel free to fork the repository and submit pull requests.
This project is licensed under the MIT License.