DualCFGL: dual-channel fusion global and local features for sequential recommendation
Abstract Sequential recommendation systems capture the dynamic interests of users and predict their future preferences. A noteworthy problem in sequential recommendation is coping with the intrinsic changes of user interests. The sequence of user interactions is generated by more than a single and s...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2024-12-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-024-01734-3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571178911268864 |
|---|---|
| author | Shuxu Chen Yuanyuan Liu Chao Che Ziqi Wei Zhaoqian Zhong |
| author_facet | Shuxu Chen Yuanyuan Liu Chao Che Ziqi Wei Zhaoqian Zhong |
| author_sort | Shuxu Chen |
| collection | DOAJ |
| description | Abstract Sequential recommendation systems capture the dynamic interests of users and predict their future preferences. A noteworthy problem in sequential recommendation is coping with the intrinsic changes of user interests. The sequence of user interactions is generated by more than a single and stable global preference, users may have interest drift that occur in a short period of time. We call this short-term interest drift as the local preference of users, which is often a key factor affecting the final choice of users. However, existing methods have limitations in observing local preferences, which leads to an incomplete consideration of the local preferences. Moreover, using a single model to represent global–local preferences obscure the distinct features of each, limiting the potential synergistic benefits. To alleviate the above limitations, we propose a novel model with a dual-channel structure to monitor both global and local preferences and ensure they complement each other. The model extracts the global preferences of users with a bidirectional Transformer using random masking and a sliding window, and extracts the local preferences with a patch-based stacked bottleneck residual convolution. To enable the model to consider both the global and local preferences of users, we design an adaptive orthogonal fusion module, which effectively fuses the two preferences and enables the two feature types to complement and enhance each other. We integrate the fused user preferences with a knowledge distillation method that further improves the model’s expressive ability. We conduct extensive experiments on three widely used datasets, and the results show that our model outperforms current state-of-the-art models. |
| format | Article |
| id | doaj-art-8472575b15984b98b84f818c17322ed8 |
| institution | Kabale University |
| issn | 2199-4536 2198-6053 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj-art-8472575b15984b98b84f818c17322ed82025-02-02T12:49:26ZengSpringerComplex & Intelligent Systems2199-45362198-60532024-12-0111111810.1007/s40747-024-01734-3DualCFGL: dual-channel fusion global and local features for sequential recommendationShuxu Chen0Yuanyuan Liu1Chao Che2Ziqi Wei3Zhaoqian Zhong4Key Laboratory of Advanced Design and Intelligent Computing, Ministry of Education, Dalian UniversityBiomedical Big Data Center, The First Affiliated Hospital, Zhejiang UniversitySchool of Software Engineering, Dalian UniversityInstitute of Automation, Chinese Academy of SciencesKey Laboratory of Advanced Design and Intelligent Computing, Ministry of Education, Dalian UniversityAbstract Sequential recommendation systems capture the dynamic interests of users and predict their future preferences. A noteworthy problem in sequential recommendation is coping with the intrinsic changes of user interests. The sequence of user interactions is generated by more than a single and stable global preference, users may have interest drift that occur in a short period of time. We call this short-term interest drift as the local preference of users, which is often a key factor affecting the final choice of users. However, existing methods have limitations in observing local preferences, which leads to an incomplete consideration of the local preferences. Moreover, using a single model to represent global–local preferences obscure the distinct features of each, limiting the potential synergistic benefits. To alleviate the above limitations, we propose a novel model with a dual-channel structure to monitor both global and local preferences and ensure they complement each other. The model extracts the global preferences of users with a bidirectional Transformer using random masking and a sliding window, and extracts the local preferences with a patch-based stacked bottleneck residual convolution. To enable the model to consider both the global and local preferences of users, we design an adaptive orthogonal fusion module, which effectively fuses the two preferences and enables the two feature types to complement and enhance each other. We integrate the fused user preferences with a knowledge distillation method that further improves the model’s expressive ability. We conduct extensive experiments on three widely used datasets, and the results show that our model outperforms current state-of-the-art models.https://doi.org/10.1007/s40747-024-01734-3Sequential recommendationGlobal and local preferenceDual-channel structureAdaptive orthogonal fusion |
| spellingShingle | Shuxu Chen Yuanyuan Liu Chao Che Ziqi Wei Zhaoqian Zhong DualCFGL: dual-channel fusion global and local features for sequential recommendation Complex & Intelligent Systems Sequential recommendation Global and local preference Dual-channel structure Adaptive orthogonal fusion |
| title | DualCFGL: dual-channel fusion global and local features for sequential recommendation |
| title_full | DualCFGL: dual-channel fusion global and local features for sequential recommendation |
| title_fullStr | DualCFGL: dual-channel fusion global and local features for sequential recommendation |
| title_full_unstemmed | DualCFGL: dual-channel fusion global and local features for sequential recommendation |
| title_short | DualCFGL: dual-channel fusion global and local features for sequential recommendation |
| title_sort | dualcfgl dual channel fusion global and local features for sequential recommendation |
| topic | Sequential recommendation Global and local preference Dual-channel structure Adaptive orthogonal fusion |
| url | https://doi.org/10.1007/s40747-024-01734-3 |
| work_keys_str_mv | AT shuxuchen dualcfgldualchannelfusionglobalandlocalfeaturesforsequentialrecommendation AT yuanyuanliu dualcfgldualchannelfusionglobalandlocalfeaturesforsequentialrecommendation AT chaoche dualcfgldualchannelfusionglobalandlocalfeaturesforsequentialrecommendation AT ziqiwei dualcfgldualchannelfusionglobalandlocalfeaturesforsequentialrecommendation AT zhaoqianzhong dualcfgldualchannelfusionglobalandlocalfeaturesforsequentialrecommendation |