《计算机应用》唯一官方网站 ›› 2025, Vol. 45 ›› Issue (4): 1223-1231.DOI: 10.11772/j.issn.1001-9081.2024040461
收稿日期:
2024-04-18
修回日期:
2024-06-12
接受日期:
2024-06-19
发布日期:
2025-04-08
出版日期:
2025-04-10
通讯作者:
宋楚君
作者简介:
党伟超(1974—),男,山西运城人,副教授,博士,CCF会员,主要研究方向:智能计算、软件可靠性基金资助:
Weichao DANG, Chujun SONG(), Gaimei GAO, Chunxia LIU
Received:
2024-04-18
Revised:
2024-06-12
Accepted:
2024-06-19
Online:
2025-04-08
Published:
2025-04-10
Contact:
Chujun SONG
About author:
DANG Weichao, born in 1974, Ph. D., associate professor. His research interests include intelligent computing, software reliability.Supported by:
摘要:
针对多行为推荐研究中存在的数据稀疏和忽视多行为之间复杂联系的问题,提出一种基于级联残差图卷积网络的多行为推荐(CRMBR)模型。首先,从由所有行为的相互作用构建的统一同构图中学习用户和项目的全局嵌入,并将这些嵌入用作初始化嵌入;其次,通过级联残差块捕获不同行为之间的联系,以不断细化不同类型行为的嵌入,从而完善用户偏好;最后,通过2种不同的聚合策略分别聚合用户和项目嵌入,并采用多任务学习(MTL)优化这些嵌入。在多个真实数据集上的实验结果表明,CRMBR模型的推荐性能优于目前的主流模型。与先进的基准模型——多行为分层图卷积网络(MB-HGCN)相比,在Tmall数据集上,所提模型的命中率(HR@20)和归一化折损累积增益(NDCG@20)分别提升了3.1%和3.9%;在Beibei数据集上,则分别提升了15.8%和16.9%;在Jdata数据集上,则分别提升了1.0%和3.3%,验证了所提模型的有效性。
中图分类号:
党伟超, 宋楚君, 高改梅, 刘春霞. 基于级联残差图卷积网络的多行为推荐[J]. 计算机应用, 2025, 45(4): 1223-1231.
Weichao DANG, Chujun SONG, Gaimei GAO, Chunxia LIU. Multi-behavior recommendation based on cascading residual graph convolutional network[J]. Journal of Computer Applications, 2025, 45(4): 1223-1231.
数据集 | 用户数 | 项目数 | 购买数 | 加购数 | 收藏数 | 浏览数 |
---|---|---|---|---|---|---|
Tmall | 41 738 | 11 953 | 255 586 | 1 996 | 221 514 | 1 813 498 |
Beibei | 21 716 | 7 997 | 304 576 | 642 622 | — | 2 412 586 |
Jdata | 93 334 | 24 624 | 333 383 | 49 891 | 45 613 | 1 681 430 |
表1 数据集统计信息
Tab. 1 Statistical information of datasets
数据集 | 用户数 | 项目数 | 购买数 | 加购数 | 收藏数 | 浏览数 |
---|---|---|---|---|---|---|
Tmall | 41 738 | 11 953 | 255 586 | 1 996 | 221 514 | 1 813 498 |
Beibei | 21 716 | 7 997 | 304 576 | 642 622 | — | 2 412 586 |
Jdata | 93 334 | 24 624 | 333 383 | 49 891 | 45 613 | 1 681 430 |
数据集 | 评价指标 | 单行为 | 多行为 | 提高/% | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
MF-BPR | NCF | LightGCN | R-GCN | NMTR | MBGCN | GNMR | S-MBRec | CRGCN | MB-HGCN | CRMBR | |||
Tmall | HR@10 | 0.023 0 | 0.030 1 | 0.039 3 | 0.031 6 | 0.051 7 | 0.054 9 | 0.039 3 | 0.069 4 | 0.084 0 | 0.150 2 | 2.8 | |
NDCG@10 | 0.012 4 | 0.015 3 | 0.020 9 | 0.015 7 | 0.025 0 | 0.028 5 | 0.019 3 | 0.036 2 | 0.044 2 | 0.080 1 | 4.0 | ||
HR@20 | 0.031 6 | 0.042 0 | 0.053 8 | 0.048 9 | 0.084 7 | 0.079 9 | 0.061 9 | 0.100 9 | 0.123 8 | 0.213 6 | 3.1 | ||
NDCG@20 | 0.014 4 | 0.018 2 | 0.024 3 | 0.019 8 | 0.033 0 | 0.034 5 | 0.024 7 | 0.043 8 | 0.054 0 | 0.095 6 | 3.9 | ||
HR@50 | 0.043 4 | 0.067 8 | 0.081 3 | 0.082 6 | 0.149 8 | 0.128 5 | 0.107 1 | 0.155 3 | 0.199 4 | 0.318 6 | 1.2 | ||
NDCG@50 | 0.016 6 | 0.023 1 | 0.029 5 | 0.026 2 | 0.045 6 | 0.043 8 | 0.033 2 | 0.054 4 | 0.068 5 | 0.116 0 | 2.7 | ||
Beibei | HR@10 | 0.026 8 | 0.029 6 | 0.030 9 | 0.032 7 | 0.031 5 | 0.037 3 | 0.039 6 | 0.048 9 | 0.053 9 | 0.069 4 | 12.1 | |
NDCG@10 | 0.013 9 | 0.014 6 | 0.016 1 | 0.016 1 | 0.014 6 | 0.019 3 | 0.021 9 | 0.025 3 | 0.025 9 | 0.034 3 | 15.5 | ||
HR@20 | 0.042 7 | 0.045 3 | 0.047 8 | 0.056 1 | 0.058 7 | 0.063 9 | 0.064 0 | 0.077 0 | 0.094 4 | 0.118 0 | 15.8 | ||
NDCG@20 | 0.017 9 | 0.018 5 | 0.020 4 | 0.021 9 | 0.021 4 | 0.025 9 | 0.028 0 | 0.032 4 | 0.036 1 | 0.046 4 | 16.9 | ||
HR@50 | 0.079 3 | 0.080 9 | 0.088 0 | 0.111 8 | 0.127 6 | 0.128 7 | 0.121 9 | 0.123 4 | 0.181 7 | 0.229 7 | 14.3 | ||
NDCG@50 | 0.025 0 | 0.021 6 | 0.028 2 | 0.032 9 | 0.034 8 | 0.038 6 | 0.039 4 | 0.041 5 | 0.053 2 | 0.068 4 | 15.5 | ||
Jdata | HR@10 | 0.185 0 | 0.209 0 | 0.225 2 | 0.240 6 | 0.314 2 | 0.280 3 | 0.306 8 | 0.412 5 | 0.500 1 | 0.541 1 | 1.4 | |
NDCG@10 | 0.123 8 | 0.141 0 | 0.143 6 | 0.144 4 | 0.171 7 | 0.157 2 | 0.158 1 | 0.277 9 | 0.291 4 | 0.329 2 | 1.7 | ||
HR@20 | 0.219 2 | 0.246 1 | 0.282 5 | 0.341 8 | 0.408 6 | 0.360 3 | 0.369 4 | 0.495 7 | 0.619 0 | 0.651 5 | 1.0 | ||
NDCG@20 | 0.132 5 | 0.150 4 | 0.158 2 | 0.158 8 | 0.196 6 | 0.179 0 | 0.194 4 | 0.298 9 | 0.322 5 | 0.364 8 | 3.3 | ||
HR@50 | 0.265 2 | 0.293 4 | 0.365 8 | 0.487 3 | 0.522 7 | 0.504 5 | 0.460 7 | 0.603 6 | 0.768 5 | 0.783 6 | 1.1 | ||
NDCG@50 | 0.141 7 | 0.159 9 | 0.174 7 | 0.189 1 | 0.219 8 | 0.198 4 | 0.202 9 | 0.320 3 | 0.353 5 | 0.390 2 | 2.6 |
表2 整体性能比较结果
Tab. 2 Overall performance comparison results
数据集 | 评价指标 | 单行为 | 多行为 | 提高/% | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
MF-BPR | NCF | LightGCN | R-GCN | NMTR | MBGCN | GNMR | S-MBRec | CRGCN | MB-HGCN | CRMBR | |||
Tmall | HR@10 | 0.023 0 | 0.030 1 | 0.039 3 | 0.031 6 | 0.051 7 | 0.054 9 | 0.039 3 | 0.069 4 | 0.084 0 | 0.150 2 | 2.8 | |
NDCG@10 | 0.012 4 | 0.015 3 | 0.020 9 | 0.015 7 | 0.025 0 | 0.028 5 | 0.019 3 | 0.036 2 | 0.044 2 | 0.080 1 | 4.0 | ||
HR@20 | 0.031 6 | 0.042 0 | 0.053 8 | 0.048 9 | 0.084 7 | 0.079 9 | 0.061 9 | 0.100 9 | 0.123 8 | 0.213 6 | 3.1 | ||
NDCG@20 | 0.014 4 | 0.018 2 | 0.024 3 | 0.019 8 | 0.033 0 | 0.034 5 | 0.024 7 | 0.043 8 | 0.054 0 | 0.095 6 | 3.9 | ||
HR@50 | 0.043 4 | 0.067 8 | 0.081 3 | 0.082 6 | 0.149 8 | 0.128 5 | 0.107 1 | 0.155 3 | 0.199 4 | 0.318 6 | 1.2 | ||
NDCG@50 | 0.016 6 | 0.023 1 | 0.029 5 | 0.026 2 | 0.045 6 | 0.043 8 | 0.033 2 | 0.054 4 | 0.068 5 | 0.116 0 | 2.7 | ||
Beibei | HR@10 | 0.026 8 | 0.029 6 | 0.030 9 | 0.032 7 | 0.031 5 | 0.037 3 | 0.039 6 | 0.048 9 | 0.053 9 | 0.069 4 | 12.1 | |
NDCG@10 | 0.013 9 | 0.014 6 | 0.016 1 | 0.016 1 | 0.014 6 | 0.019 3 | 0.021 9 | 0.025 3 | 0.025 9 | 0.034 3 | 15.5 | ||
HR@20 | 0.042 7 | 0.045 3 | 0.047 8 | 0.056 1 | 0.058 7 | 0.063 9 | 0.064 0 | 0.077 0 | 0.094 4 | 0.118 0 | 15.8 | ||
NDCG@20 | 0.017 9 | 0.018 5 | 0.020 4 | 0.021 9 | 0.021 4 | 0.025 9 | 0.028 0 | 0.032 4 | 0.036 1 | 0.046 4 | 16.9 | ||
HR@50 | 0.079 3 | 0.080 9 | 0.088 0 | 0.111 8 | 0.127 6 | 0.128 7 | 0.121 9 | 0.123 4 | 0.181 7 | 0.229 7 | 14.3 | ||
NDCG@50 | 0.025 0 | 0.021 6 | 0.028 2 | 0.032 9 | 0.034 8 | 0.038 6 | 0.039 4 | 0.041 5 | 0.053 2 | 0.068 4 | 15.5 | ||
Jdata | HR@10 | 0.185 0 | 0.209 0 | 0.225 2 | 0.240 6 | 0.314 2 | 0.280 3 | 0.306 8 | 0.412 5 | 0.500 1 | 0.541 1 | 1.4 | |
NDCG@10 | 0.123 8 | 0.141 0 | 0.143 6 | 0.144 4 | 0.171 7 | 0.157 2 | 0.158 1 | 0.277 9 | 0.291 4 | 0.329 2 | 1.7 | ||
HR@20 | 0.219 2 | 0.246 1 | 0.282 5 | 0.341 8 | 0.408 6 | 0.360 3 | 0.369 4 | 0.495 7 | 0.619 0 | 0.651 5 | 1.0 | ||
NDCG@20 | 0.132 5 | 0.150 4 | 0.158 2 | 0.158 8 | 0.196 6 | 0.179 0 | 0.194 4 | 0.298 9 | 0.322 5 | 0.364 8 | 3.3 | ||
HR@50 | 0.265 2 | 0.293 4 | 0.365 8 | 0.487 3 | 0.522 7 | 0.504 5 | 0.460 7 | 0.603 6 | 0.768 5 | 0.783 6 | 1.1 | ||
NDCG@50 | 0.141 7 | 0.159 9 | 0.174 7 | 0.189 1 | 0.219 8 | 0.198 4 | 0.202 9 | 0.320 3 | 0.353 5 | 0.390 2 | 2.6 |
SC | Tmall | Beibei | Jdata | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
HR@10 | NDCG@10 | HR@20 | NDCG@20 | HR@10 | NDCG@10 | HR@20 | NDCG@20 | HR@10 | NDCG@10 | HR@20 | NDCG@20 | ||
√ | 0.025 8 | 0.011 7 | 0.039 4 | 0.017 2 | 0.014 9 | 0.008 4 | 0.033 7 | 0.010 6 | 0.163 9 | 0.089 2 | 0.220 1 | 0.102 8 | |
√ | 0.119 4 | 0.062 2 | 0.170 3 | 0.074 7 | 0.042 8 | 0.018 1 | 0.077 2 | 0.025 5 | 0.445 7 | 0.237 8 | 0.517 1 | 0.275 1 | |
√ | √ | 0.150 2 | 0.080 1 | 0.213 6 | 0.095 6 | 0.069 4 | 0.034 3 | 0.118 0 | 0.046 4 | 0.541 1 | 0.329 2 | 0.651 5 | 0.364 8 |
表3 不同设计对残差块的影响
Tab. 3 Influence of different designs on residual block
SC | Tmall | Beibei | Jdata | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
HR@10 | NDCG@10 | HR@20 | NDCG@20 | HR@10 | NDCG@10 | HR@20 | NDCG@20 | HR@10 | NDCG@10 | HR@20 | NDCG@20 | ||
√ | 0.025 8 | 0.011 7 | 0.039 4 | 0.017 2 | 0.014 9 | 0.008 4 | 0.033 7 | 0.010 6 | 0.163 9 | 0.089 2 | 0.220 1 | 0.102 8 | |
√ | 0.119 4 | 0.062 2 | 0.170 3 | 0.074 7 | 0.042 8 | 0.018 1 | 0.077 2 | 0.025 5 | 0.445 7 | 0.237 8 | 0.517 1 | 0.275 1 | |
√ | √ | 0.150 2 | 0.080 1 | 0.213 6 | 0.095 6 | 0.069 4 | 0.034 3 | 0.118 0 | 0.046 4 | 0.541 1 | 0.329 2 | 0.651 5 | 0.364 8 |
方法 | Tmall | Beibei | Jdata | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
HR@10 | NDCG@10 | HR@20 | NDCG@20 | HR@10 | NDCG@10 | HR@20 | NDCG@20 | HR@10 | NDCG@10 | HR@20 | NDCG@20 | |
0.102 8 | 0.053 5 | 0.151 6 | 0.065 6 | 0.030 7 | 0.015 1 | 0.076 5 | 0.019 4 | 0.335 1 | 0.193 4 | 0.356 2 | 0.229 7 | |
0.150 2 | 0.080 1 | 0.213 6 | 0.095 6 | 0.069 4 | 0.034 3 | 0.118 0 | 0.046 4 | 0.541 1 | 0.329 2 | 0.651 5 | 0.364 8 |
表4 全局图G嵌入学习的影响
Tab. 4 Influence of embedding learning in global graph G
方法 | Tmall | Beibei | Jdata | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
HR@10 | NDCG@10 | HR@20 | NDCG@20 | HR@10 | NDCG@10 | HR@20 | NDCG@20 | HR@10 | NDCG@10 | HR@20 | NDCG@20 | |
0.102 8 | 0.053 5 | 0.151 6 | 0.065 6 | 0.030 7 | 0.015 1 | 0.076 5 | 0.019 4 | 0.335 1 | 0.193 4 | 0.356 2 | 0.229 7 | |
0.150 2 | 0.080 1 | 0.213 6 | 0.095 6 | 0.069 4 | 0.034 3 | 0.118 0 | 0.046 4 | 0.541 1 | 0.329 2 | 0.651 5 | 0.364 8 |
方法 | Tmall | Beibei | Jdata | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
HR@10 | NDCG@10 | HR@20 | NDCG@20 | HR@10 | NDCG@10 | HR@20 | NDCG@20 | HR@10 | NDCG@10 | HR@20 | NDCG@20 | |
0.108 4 | 0.065 8 | 0.168 2 | 0.071 5 | 0.552 7 | 0.027 3 | 0.101 9 | 0.031 7 | 0.447 0 | 0.259 9 | 0.583 3 | 0.307 0 | |
0.135 3 | 0.071 9 | 0.184 3 | 0.082 2 | 0.060 2 | 0.029 7 | 0.107 8 | 0.037 9 | 0.519 3 | 0.302 5 | 0.607 8 | 0.346 3 | |
0.150 2 | 0.080 1 | 0.213 6 | 0.095 6 | 0.069 4 | 0.034 3 | 0.118 0 | 0.046 4 | 0.541 1 | 0.329 2 | 0.649 5 | 0.364 8 | |
0.135 9 | 0.073 6 | 0.193 2 | 0.077 5 | 0.061 1 | 0.029 3 | 0.082 7 | 0.032 9 | 0.485 6 | 0.289 1 | 0.581 7 | 0.298 1 | |
0.145 3 | 0.078 3 | 0.202 8 | 0.083 1 | 0.063 0 | 0.031 2 | 0.097 2 | 0.038 2 | 0.503 2 | 0.318 0 | 0.598 2 | 0.345 0 | |
0.150 2 | 0.080 1 | 0.213 6 | 0.095 6 | 0.069 4 | 0.034 3 | 0.118 0 | 0.046 4 | 0.541 1 | 0.329 2 | 0.651 5 | 0.364 8 |
表5 聚合策略的影响
Tab. 5 Influence of aggregation strategies
方法 | Tmall | Beibei | Jdata | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
HR@10 | NDCG@10 | HR@20 | NDCG@20 | HR@10 | NDCG@10 | HR@20 | NDCG@20 | HR@10 | NDCG@10 | HR@20 | NDCG@20 | |
0.108 4 | 0.065 8 | 0.168 2 | 0.071 5 | 0.552 7 | 0.027 3 | 0.101 9 | 0.031 7 | 0.447 0 | 0.259 9 | 0.583 3 | 0.307 0 | |
0.135 3 | 0.071 9 | 0.184 3 | 0.082 2 | 0.060 2 | 0.029 7 | 0.107 8 | 0.037 9 | 0.519 3 | 0.302 5 | 0.607 8 | 0.346 3 | |
0.150 2 | 0.080 1 | 0.213 6 | 0.095 6 | 0.069 4 | 0.034 3 | 0.118 0 | 0.046 4 | 0.541 1 | 0.329 2 | 0.649 5 | 0.364 8 | |
0.135 9 | 0.073 6 | 0.193 2 | 0.077 5 | 0.061 1 | 0.029 3 | 0.082 7 | 0.032 9 | 0.485 6 | 0.289 1 | 0.581 7 | 0.298 1 | |
0.145 3 | 0.078 3 | 0.202 8 | 0.083 1 | 0.063 0 | 0.031 2 | 0.097 2 | 0.038 2 | 0.503 2 | 0.318 0 | 0.598 2 | 0.345 0 | |
0.150 2 | 0.080 1 | 0.213 6 | 0.095 6 | 0.069 4 | 0.034 3 | 0.118 0 | 0.046 4 | 0.541 1 | 0.329 2 | 0.651 5 | 0.364 8 |
1 | ZHANG S, YAO L, SUN A, et al. Deep learning based recommender system: a survey and new perspectives[J]. ACM Computing Surveys, 2019, 52(1): No.5. |
2 | XU Y, ZHU L, CHENG Z, et al. Multi-feature discrete collaborative filtering for fast cold-start recommendation[C]// Proceedings of the 34th AAAI Conference on Artificial Intelligence. Palo Alto: AAAI Press, 2020:270-278. |
3 | KOREN Y, BELL R, VOLINSKY C. Matrix factorization techniques for recommender systems[J]. Computer, 2009, 42(8):30-37. |
4 | 曹健,孙浩,李海生,等. 基于深度神经网络的会话推荐研究[J]. 计算机仿真, 2023, 40(10):1-8. |
CAO J, SUN H, LI H S, et al. Deep neural networks in session-based recommendation: a survey[J]. Computer Simulation, 2023, 40(10):1-8. | |
5 | CHEN Y, TANG Y, YUAN Y. Attention-enhanced graph neural networks with global context for session-based recommendation[J]. IEEE Access, 2023, 11:26237-26246. |
6 | HE X, LIAO L, ZHANG H, et al. Neural collaborative filtering[C]// Proceedings of the 26th International Conference on World Wide Web. Republic and Canton of Geneva: International World Wide Web Conferences Steering Committee, 2017: 173-182. |
7 | PENG H, ZHANG R, DOU Y, et al. Reinforced neighborhood selection guided multi-relational graph neural networks[J]. ACM Transactions on Information Systems, 2022, 40(4): No.69. |
8 | 严明时,程志勇,孙静,等. 基于两阶段学习的多行为推荐[J]. 软件学报, 2024, 35(5):2446-2465. |
YAN M S, CHENG Z Y, SUN J, et al. Two-stage learning for multi-behavior recommendation[J]. Journal of Software, 2024, 35(5):2446-2465. | |
9 | SINGH A P, GORDON G J. Relational learning via collective matrix factorization[C]// Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2008:650-658. |
10 | ZHAO Z, CHENG Z, HONG L, et al. Improving user topic interest profiles by behavior factorization[C]// Proceedings of the 24th International Conference on World Wide Web. Republic and Canton of Geneva: International World Wide Web Conferences Steering Committee, 2015:1406-1416. |
11 | LONI B, PAGANO R, LARSON M, et al. Bayesian personalized ranking with multi-channel user feedback[C]// Proceedings of the 10th ACM Conference on Recommender Systems. New York: ACM, 2016: 361-364. |
12 | DING J, YU G, HE X, et al. Improving implicit recommender systems with view data[C]// Proceedings of the 27th International Joint Conference on Artificial Intelligence. California: ijcai.org, 2018: 3343-3349. |
13 | GUO G, QIU H, TAN Z, et al. Resolving data sparsity by multi-type auxiliary implicit feedback for recommender systems[J]. Knowledge-Based Systems, 2017, 138:202-207. |
14 | QIU H, LIU Y, GUO G, et al. BPRH: Bayesian personalized ranking for heterogeneous implicit feedback[J]. Information Sciences, 2018, 453: 80-98. |
15 | DONG Y, JIANG W. Brand purchase prediction based on time-evolving user behaviors in e-commerce[J]. Concurrency and Computation: Practice and Experience, 2019, 31(1): No.e4882. |
16 | XIA L, HUANG C, XU Y, et al. Multiplex behavioral relation learning for recommendation via memory augmented transformer network[C]// Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2020:2397-2406. |
17 | GUO L, HUA L, JIA R, et al. Buying or browsing? predicting real-time purchasing intent using attention-based deep network with multiple behavior[C]// Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2019:1984-1992. |
18 | GAO G, HE X, GAN D, et al. Learning to recommend with multiple cascading behaviors[J]. IEEE Transactions on Knowledge and Data Engineering, 2021, 33(6): 2588-2601. |
19 | 李驰,游小钰,张谧. 基于解耦图卷积网络的协同过滤推荐模型[J]. 中文信息学报, 2023, 37(11):131-141. |
LI C, YOU X Y, ZHANG M. Decoupled graph convolution network for collaborative filtering[J]. Journal of Chinese Information Processing, 2023, 37(11):131-141. | |
20 | YAN M, CHENG Z, SUN J, et al. MB-HGCN: a hierarchical graph convolutional network for multi-behavior recommendation[EB/OL]. [2024-02-20].. |
21 | WANG X, HE X, WANG M, et al. Neural graph collaborative filtering[C]// Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2019:165-174. |
22 | HE X, DENG K, WANG X, et al. LightGCN: simplifying and powering graph convolution network for recommendation[C]// Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2020:639-648. |
23 | WANG X, JIN H, ZHANG A, et al. Disentangled graph collaborative filtering[C]// Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2020:1001-1010. |
24 | JIN B, GAO C, HE X, et al. Multi-behavior recommendation with graph convolutional networks[C]// Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2020:659-668. |
25 | XIA L, HUANG C, XU Y, et al. Multi-behavior enhanced recommendation with cross-interaction collaborative relation modeling[C]// Proceedings of the IEEE 37th International Conference on Data Engineering. Piscataway: IEEE, 2021:1931-1936. |
26 | ZHANG W, MAO J, CAO Y, et al. Multiplex graph neural networks for multi-behavior recommendation[C]// Proceedings of the 29th ACM International Conference on Information and Knowledge Management. New York: ACM, 2020:2313-2316. |
27 | YAN M, CHENG Z, GAO C, et al. Cascading residual graph convolutional network for multi-behavior recommendation[J]. ACM Transactions on Information Systems, 2024, 42(1): No.10. |
28 | 曹磊亮,谢瑾奎,郭天晟. 结合残差网络和噪声处理的节点分类模型[J]. 小型微型计算机系统, 2024, 45(6):1331-1338. |
CAO L L, XIE J K, GUO T S. Node classification models combining residual networks and noise processing[J]. Journal of Chinese Computer Systems, 2024, 45(6):1331-1338. | |
29 | 唐宇,吴贞东. 基于残差网络的轻量级图卷积推荐方法[J]. 计算机工程与应用, 2024, 60(3):205-212. |
TANG Y, WU Z D. Light graph convolution recommendation method based on residual network[J]. Computer Engineering and Applications, 2024, 60(3):205-212. | |
30 | TANG H, LIU J, ZHAO M, et al. Progressive Layered Extraction (PLE): a novel Multi-Task Learning (MTL) model for personalized recommendations[C]// Proceedings of the 14th ACM Conference on Recommender Systems. New York: ACM, 2020:269-278. |
31 | RENDLE S, FREUDENTHALER C, GANTNER Z, et al. BPR: Bayesian personalized ranking from implicit feedback[C]// Proceedings of the 25th International Conference on Uncertainty in Artificial Intelligence. Arlington, VA: AUAI Press, 2009:452-461. |
32 | SCHLICHTKRULL M, KIPF T N, BLOEM P, et al. Modeling relational data with graph convolutional networks[C]// Proceedings of the 2018 European Semantic Web Conference, LNCS 10843. Cham: Springer, 2018:593-607. |
33 | GU S, WANG X, SHI C, et al. Self-supervised graph neural networks for multi-behavior recommendation[C]// Proceedings of the 31st International Joint Conference on Artificial Intelligence. California: ijcai.org, 2022: 2052-2058. |
[1] | 富坤, 应世聪, 郑婷婷, 屈佳捷, 崔静远, 李建伟. 面向小样本节点分类的图数据增强方法[J]. 《计算机应用》唯一官方网站, 2025, 45(2): 392-402. |
[2] | 徐欣然, 张绍兵, 成苗, 张洋, 曾尚. 基于多路层次化混合专家模型的轴承故障诊断方法[J]. 《计算机应用》唯一官方网站, 2025, 45(1): 59-68. |
[3] | 薛桂香, 王辉, 周卫峰, 刘瑜, 李岩. 基于知识图谱和时空扩散图卷积网络的港口交通流量预测[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2952-2957. |
[4] | 庞川林, 唐睿, 张睿智, 刘川, 刘佳, 岳士博. D2D通信系统中基于图卷积网络的分布式功率控制算法[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2855-2862. |
[5] | 刘禹含, 吉根林, 张红苹. 基于骨架图与混合注意力的视频行人异常检测方法[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2551-2557. |
[6] | 李欢欢, 黄添强, 丁雪梅, 罗海峰, 黄丽清. 基于多尺度时空图卷积网络的交通出行需求预测[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2065-2072. |
[7] | 吕锡婷, 赵敬华, 荣海迎, 赵嘉乐. 基于Transformer和关系图卷积网络的信息传播预测模型[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1760-1766. |
[8] | 黎施彬, 龚俊, 汤圣君. 基于Graph Transformer的半监督异配图表示学习模型[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1816-1823. |
[9] | 李威, 陈玲, 徐修远, 朱敏, 郭际香, 周凯, 牛颢, 张煜宸, 易珊烨, 章毅, 罗凤鸣. 基于多任务学习的间质性肺病分割算法[J]. 《计算机应用》唯一官方网站, 2024, 44(4): 1285-1293. |
[10] | 高龙涛, 李娜娜. 基于方面感知注意力增强的方面情感三元组抽取[J]. 《计算机应用》唯一官方网站, 2024, 44(4): 1049-1057. |
[11] | 杨先凤, 汤依磊, 李自强. 基于交替注意力机制和图卷积网络的方面级情感分析模型[J]. 《计算机应用》唯一官方网站, 2024, 44(4): 1058-1064. |
[12] | 尚爱国, 朱欣娟. 基于多任务学习的意图检测和槽位填充联合方法[J]. 《计算机应用》唯一官方网站, 2024, 44(3): 690-695. |
[13] | 王楷天, 叶青, 程春雷. 基于异构图表示的中医电子病历分类方法[J]. 《计算机应用》唯一官方网站, 2024, 44(2): 411-417. |
[14] | 宋钰丹, 王晶, 王雪徽, 马朝阳, 林友芳. 基于自适应多任务学习的睡眠生理时序分类方法[J]. 《计算机应用》唯一官方网站, 2024, 44(2): 654-662. |
[15] | 廖存燚, 郑毅, 刘玮瑾, 于欢, 刘守印. 自动驾驶环境感知多任务去耦-融合算法[J]. 《计算机应用》唯一官方网站, 2024, 44(2): 424-431. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||