Modeling Stage-wise Evolution of User Interests for News Recommendation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cheng, Zhiyong, Jin, Yike, Zhang, Zhijie, Chen, Huilin, Duan, Zhangling, Wang, Meng
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911505671258112
author Cheng, Zhiyong
Jin, Yike
Zhang, Zhijie
Chen, Huilin
Duan, Zhangling
Wang, Meng
author_facet Cheng, Zhiyong
Jin, Yike
Zhang, Zhijie
Chen, Huilin
Duan, Zhangling
Wang, Meng
contents Personalized news recommendation is highly time-sensitive, as user interests are often driven by emerging events, trending topics, and shifting real-world contexts. These dynamics make it essential to model not only users' long-term preferences, which reflect stable reading habits and high-order collaborative patterns, but also their short-term, context-dependent interests that change rapidly over time. However, most existing approaches rely on a single static interaction graph, which struggles to capture both long-term preference patterns and short-term interest changes as user behavior evolves. To address this challenge, we propose a unified framework that learns user preferences from both global and local temporal perspectives. A global preference modeling component captures long-term collaborative signals from the overall interaction graph, while a local preference modeling component partitions historical interactions into stage-wise temporal subgraphs to represent short-term dynamics. Within this module, an LSTM branch models the progressive evolution of recent interests, and a self-attention branch captures long-range temporal dependencies. Extensive experiments on two large-scale real-world datasets show that our approach consistently outperforms strong baselines and delivers fresher and more relevant recommendations across diverse user behaviors and temporal settings.
format Preprint
id arxiv_https___arxiv_org_abs_2603_10471
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Modeling Stage-wise Evolution of User Interests for News Recommendation
Cheng, Zhiyong
Jin, Yike
Zhang, Zhijie
Chen, Huilin
Duan, Zhangling
Wang, Meng
Information Retrieval
Artificial Intelligence
H.3.3; H.5.1
Personalized news recommendation is highly time-sensitive, as user interests are often driven by emerging events, trending topics, and shifting real-world contexts. These dynamics make it essential to model not only users' long-term preferences, which reflect stable reading habits and high-order collaborative patterns, but also their short-term, context-dependent interests that change rapidly over time. However, most existing approaches rely on a single static interaction graph, which struggles to capture both long-term preference patterns and short-term interest changes as user behavior evolves. To address this challenge, we propose a unified framework that learns user preferences from both global and local temporal perspectives. A global preference modeling component captures long-term collaborative signals from the overall interaction graph, while a local preference modeling component partitions historical interactions into stage-wise temporal subgraphs to represent short-term dynamics. Within this module, an LSTM branch models the progressive evolution of recent interests, and a self-attention branch captures long-range temporal dependencies. Extensive experiments on two large-scale real-world datasets show that our approach consistently outperforms strong baselines and delivers fresher and more relevant recommendations across diverse user behaviors and temporal settings.
title Modeling Stage-wise Evolution of User Interests for News Recommendation
topic Information Retrieval
Artificial Intelligence
H.3.3; H.5.1
url https://arxiv.org/abs/2603.10471