AutoSeqRec: Autoencoder for Efficient Sequential Recommendation

By S. Liu et al.

Published on Oct. 21, 2023

Read the original document by opening this link in a new tab.

Table of Contents

1. Introduction
2. Preliminaries
3. The Proposed Method
3.1 Architecture
3.2 Incremental Inference

Summary

AutoSeqRec is a novel recommendation model designed for sequential recommendation tasks. It utilizes an autoencoder architecture with collaborative information, source information, and target information to capture user preferences and behaviors effectively. The model takes into account both long-term user preferences and short-term interests by reconstructing user-item interaction and item transition matrices. The incremental inference process allows for real-time recommendation updates without retraining the model. AutoSeqRec has shown superior performance in accuracy and efficiency compared to existing methods.