Time-based Sequence Model for Personalization and Recommendation Systems

TL;DR

This paper introduces a time-aware sequence model for personalized recommendations, utilizing a novel attention mechanism that effectively handles variable-length user behavior sequences and outperforms existing models on real datasets.

Contribution

The paper presents a new time-based sequence model with a TSL attention mechanism, improving recommendation accuracy and efficiency over traditional models.

Findings

Outperforms complex models on Taobao dataset

Handles variable-length sequences efficiently

Demonstrates superior performance on designed data

Abstract

In this paper we develop a novel recommendation model that explicitly incorporates time information. The model relies on an embedding layer and TSL attention-like mechanism with inner products in different vector spaces, that can be thought of as a modification of multi-headed attention. This mechanism allows the model to efficiently treat sequences of user behavior of different length. We study the properties of our state-of-the-art model on statistically designed data set. Also, we show that it outperforms more complex models with longer sequence length on the Taobao User Behavior dataset.