RSAttAE: An Information-Aware Attention-based Autoencoder Recommender System

TL;DR

This paper presents RSAttAE, an attention-based autoencoder combined with XGBoost for improved movie rating prediction, outperforming many existing methods on the MovieLens 100K dataset.

Contribution

It introduces a novel attention-based autoencoder architecture integrated with XGBoost for more accurate user-movie rating predictions.

Findings

Outperforms most state-of-the-art methods on MovieLens 100K

Utilizes attention mechanisms for better representation learning

Combines autoencoder with XGBoost for enhanced prediction accuracy

Abstract

Recommender systems play a crucial role in modern life, including information retrieval, the pharmaceutical industry, retail, and entertainment. The entertainment sector, in particular, attracts significant attention and generates substantial profits. This work proposes a new method for predicting unknown user-movie ratings to enhance customer satisfaction. To achieve this, we utilize the MovieLens 100K dataset. Our approach introduces an attention-based autoencoder to create meaningful representations and the XGBoost method for rating predictions. The results demonstrate that our proposal outperforms most of the existing state-of-the-art methods. Availability: github.com/ComputationIASBS/RecommSys