Journal article
Coformer for session-based recommendation with dual positional information
Expert systems with applications, Vol.283, 127732
07/15/2025
DOI: 10.1016/j.eswa.2025.127732
Abstract
Session-based recommendation aims to provide recommendation list based on the recent interaction behaviors of anonymous users. Existing session-based recommender systems adopt various methods to capture the behavioral patterns of anonymous users in the current session. However, most of the recommenders assumed that the type of positional information is single, neglecting the influence of other potential positional information on learning preferences. Although a few related works consider the impact of multiple positional information, they are limited in performance by inefficient fusion methods. These methods mainly based on invasive fusion which mixes different types of positional information together, fails to distinguish the various preferences and overwhelms the item representation. To address these challenges, we propose Coformer-DP which fuses user preferences hidden in general and personalized positional information based on the non-invasive method. Furthermore, the adopted contextual preference efficiently regulates the long-term preference and the comprehensive preference. Extensive experiments on four real-world datasets show the distinguished performance of Coformer-DP.
•We argue the effectiveness of considering multi-positional information.•We highlight the recurrent model to provide the implicit positional features.•The explicit positional features can be derived from position encodings.•We propose a double-branched self-attention network called Coformer.•Stable non-invasive feature fusion method efficiently utilizes positional information.
Details
- Title: Subtitle
- Coformer for session-based recommendation with dual positional information
- Creators
- Weiyue Li - Dongbei University of Finance and EconomicsZhiguo Zhu - Dongbei University of Finance and EconomicsHao Dong - Dongbei University of Finance and EconomicsCheng Chen - Dongbei University of Finance and EconomicsMing Gao - Dongbei University of Finance and EconomicsWeiguo Fan - Department of Business Analytics, Tippie College of Business, University of Iowa, United States
- Resource Type
- Journal article
- Publication Details
- Expert systems with applications, Vol.283, 127732
- Publisher
- Elsevier Ltd
- DOI
- 10.1016/j.eswa.2025.127732
- ISSN
- 0957-4174
- eISSN
- 1873-6793
- Grant note
- National Natural Science Foundation of China: 72172025, 72293563, 72101051, 72442025 Humanities and Social Sciences Foundation of the Ministry of Education of China: 21YJAZH130 Basic Scientific Research Project of Liaoning Provincial Department of Education: LJKMZ20221606, JYTZD2023050 Natural Science Foundation of Liaoning Province, China: 2024-MS-175 Liaoning Province Key Research and Development Project: 2024JH2/102400020 Dalian Scientific and Technological Talents Innovation Support Plan: 2022RG17
This research was supported and funded by the National Natural Science Foundation of China (No. 72172025, 72293563, 72101051, 72442025) ; the Humanities and Social Sciences Foundation of the Ministry of Education of China (No. 21YJAZH130) ; the Basic Scientific Research Project of Liaoning Provincial Department of Education (No. LJKMZ20221606, JYTZD2023050) ; the Natural Science Foundation of Liaoning Province, China (No. 2024-MS-175) ; the Liaoning Province Key Research and Development Project (No. 2024JH2/102400020) ; and the Dalian Scientific and Technological Talents Innovation Support Plan (No. 2022RG17) . The authors would also like to thank Hongyu Wang of the Institute of Computing Technology, Chinese Academy of Sciences, for helpful discussions related to this work.
- Language
- English
- Date published
- 07/15/2025
- Academic Unit
- Business Analytics
- Record Identifier
- 9984820564302771
Metrics
1 Record Views