GithubHelp home page GithubHelp logo

multi-aspect-control-graph-recommendation-system's Introduction

Multi-Aspect-Control-Graph-Recommendation-System

Graph neural networks (GNNs) have recently shown impressive re- sults in recommender systems, achieving remarkable results across various sectors. However, their deployment in such systems is not without challenges. One significant issue is the initialization of random weights, where incorrect starting points can hinder the learning process, causing slow convergence or getting stuck in local minima. To address this, we introduce a novel two-phase method tailored to the unique demands of recommender systems. Initially, we refine the latent weights by aligning them with be- havioral pattern similarities, inspired by the collaborative filtering technique. Subsequently, we implement a new strategy for manag- ing the number of negative samples and devising an innovative loss function. These strategic modifications substantially boost GNN performance, aligning them more closely with the needs of recom- mender systems. This is evidenced by our method outperforming existing state-of-the-art methods on two different public datasets.

Features

  • Weighted loss function to prioritize informative training samples
  • Adaptive margins to learn fine-grained user-item relationships
  • Similarity-based embedding initialization for faster convergence
  • Strategic edge removal techniques (bridge removal and noisy edge removal) to reduce computational complexity and improve embedding quality
  • Comprehensive evaluation on benchmark datasets (Amazon-Book and Gowalla)

Requirements

  • Python 3.x
  • PyTorch
  • NumPy
  • SciPy
  • scikit-learn

Usage

  1. Clone the repository:
git clone https://github.com/OrKatz7/Multi-Aspect-Control-Graph-Recommendation-System
cd MultiAspectControlGraphRecommendationSystem
  1. Prepare the dataset:
  • Download the Amazon-Book and Gowalla datasets
  • Preprocess the data and save it in the appropriate format, use preprocessing.py to initilaize the similiarty-based embedding matrix.
  1. Train:
run_amazon.sh

License

This project is licensed under the MIT License.

multi-aspect-control-graph-recommendation-system's People

Contributors

orkatz7 avatar

Watchers

 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.