Skip to content

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.

Notifications You must be signed in to change notification settings

Tamunonengiyeofori/A_I_RECOMMENDER_SYSTEM

Repository files navigation

AI-Based Recommender System for Electronic Products

Overview

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.

Dataset

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.

Methodology

Machine Learning Models

Four supervised machine learning models were employed:

  • Singular Value Decomposition (SVD)
  • Non-Negative Matrix Factorization (NMF)
  • K Nearest Neighbor (KNN)
  • Co-Clustering

Recommendation Process

  • 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.

Model Selection

The best machine learning model was selected based on performance metrics:

  • Root Mean Square Error (RMSE)
  • Mean Absolute Error (MAE)

Results

  • 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.

Usage

You can use this recommender system by following the instructions in the code provided in the project's GitHub repository: GitHub Repository.

Contributing

If you would like to contribute to this project, please follow our Contributing Guidelines.

About

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.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published