This repository contains an AI-based Recommender System for electronic products. The system recommends electronic products to users based on their preferences and past interactions. It utilizes a dataset containing information about 50 distinct electronic projects purchased by 6,493 distinct users.
The dataset used in this project can be found here. It includes information about electronic products and user interactions, making it ideal for building a recommendation system.
Four supervised machine learning models were employed:
- Singular Value Decomposition (SVD)
- Non-Negative Matrix Factorization (NMF)
- K Nearest Neighbor (KNN)
- Co-Clustering
- User-item ratings prediction was done using item-based Pearson similarity and cosine similarity.
- Models were cross-validated using 5-fold cross-validation to assess their performance.
The best machine learning model was selected based on performance metrics:
- Root Mean Square Error (RMSE)
- Mean Absolute Error (MAE)
- The SVD model demonstrated superior performance with low RMSE and MAE values, indicating high accuracy and consistency in predicting user-item ratings.
- NMF, Co-Clustering, and KNN models provided reasonable predictions but showed less consistent performance across different folds due to greater standard deviations in RMSE and MAE values.
You can use this recommender system by following the instructions in the code provided in the project's GitHub repository: GitHub Repository.
If you would like to contribute to this project, please follow our Contributing Guidelines.