{"type":"ScholarlyArticle","title":"Learning the Compositional Visual Coherence for Complementary Recommendations","authors":[{"name":"Zhi Li","slug":"zhi-li","given":"Zhi","family":"Li","affiliation":"Anhui Province Key Laboratory of Big Data Analysis and Application , School of Data Science , University of Science and Technology of China"},{"name":"Bo Wu","slug":"bo-wu","given":"Bo","family":"Wu","affiliation":"Columbia University"},{"name":"Qi Liu","slug":"qi-liu","given":"Qi","family":"Liu","affiliation":"Anhui Province Key Laboratory of Big Data Analysis and Application , School of Data Science , University of Science and Technology of China; School of Computer Science and Technology , University of Science and Technology of China"},{"name":"Likang Wu","slug":"likang-wu","given":"Likang","family":"Wu","affiliation":"School of Computer Science and Technology , University of Science and Technology of China"},{"name":"Hongke Zhao","slug":"hongke-zhao","given":"Hongke","family":"Zhao","affiliation":"The College of Management and Economics , Tianjin University"},{"name":"Tao Mei","slug":"tao-mei","given":"Tao","family":"Mei","affiliation":"JD AI Research"}],"abstract":"Complementary recommendations, which aim at providing users product suggestions that are supplementary and compatible with their obtained items, have become a hot topic in both academia and industry in recent years. Existing work mainly focused on modeling the co-purchased relations between two items, but the compositional associations of item collections are largely unexplored. Actually, when a user chooses the complementary items for the purchased products, it is intuitive that she will consider the visual semantic coherence (such as color collocations, texture compatibilities) in addition to global impressions. Towards this end, in this paper, we propose a novel Content Attentive Neural Network (CANN) to model the comprehensive compositional coherence on both global contents and semantic contents. Specifically, we first propose a Global Coherence Learning (GCL) module based on multi-heads attention to model the global compositional coherence. Then, we generate the semantic-focal representations from different semantic regions and design a Focal Coherence Learning (FCL) module to learn the focal compositional coherence from different semantic-focal representations. Finally, we optimize the CANN in a novel compositional optimization strategy. Extensive experiments on the large-scale real-world data clearly demonstrate the effectiveness of CANN compared with several state-of-the-art methods.","doi":"10.24963/ijcai.2020/489","url":"https://ijcai.org/proceedings/2020/489","pdf":"https://ijcai.org/proceedings/2020/0489.pdf","markdown":"https://ijcai.org/proceedings/2020/489.md","bibtex":"https://ijcai.org/proceedings/2020/bibtex/489","volume":{"slug":"2020","title":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2020","url":"https://ijcai.org/proceedings/2020/","year":2020,"location":"Yokohama","dates":"January 2021 (postponed from 11-17 July 2020)","editors":"Christian Bessiere","isbn":"978-0-9992411-6-5","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2020"},"section":"Multidisciplinary Topics and Applications","track":"Main track","number":489,"pages":{"first":3536,"last":3543},"keywords":["Humans and AI: Personalization and User Modeling","Data Mining: Mining Text, Web, Social Media","Multidisciplinary Topics and Applications: Recommender Systems","Machine Learning: Deep Learning"],"videos":[{"label":"Video","url":"https://videos.ijcai20.org/25101/file_25101.m3u8","length_sec":299,"provider":"hls"}],"references":[{"position":1,"raw":"[Chang et al., 2017] Y. Chang, W. Cheng, B. Wu, and K. Hua. Fashion world map: Understanding cities through streetwear fashion. In ACM MM, pages 91-99, 2017.","title":"Fashion world map: Understanding cities through streetwear fashion","authors":"Y Chang, W Cheng, B Wu, K Hua","year":"2017","venue":"ACM MM","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Cui et al., 2019] Z. Cui, Z. Li, S. Wu, X. Zhang, and L. Wang. Dressing as a whole: Outfit compatibility learn- ing based on node-wise graph neural networks. In WWW, pages 307-317. ACM, 2019.","title":"Dressing as a whole: Outfit compatibility learning based on node-wise graph neural networks","authors":"Z Cui, Z Li, S Wu, X Zhang, L Wang","year":"2019","venue":"WWW","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Deng et al., 2009] J. Deng, W. Dong, R. Socher, L. Li, K. Li, and F. Li. Imagenet: A large-scale hierarchical im- age database. In CVPR, pages 248-255, 2009.","title":"Imagenet: A large-scale hierarchical image database","authors":"J Deng, W Dong, R Socher, L Li, K Li, F Li","year":"2009","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Felzenszwalb and Huttenlocher, 2004] P F Felzenszwalb and D P Huttenlocher. Efficient graph-based image seg- mentation. IJCV, 59(2):167-181, 2004.","title":"Efficient graph-based image segmentation","authors":"P F Felzenszwalb, D P Huttenlocher","year":"2004","venue":"IJCV","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Han et al., 2017] X. Han, Z. Wu, Y. Jiang, and L. S. Davis. Learning fashion compatibility with bidirectional lstms. In ACM MM, pages 1078-1086. ACM, 2017.","title":"Learning fashion compatibility with bidirectional lstms","authors":"Han","year":"2017","venue":"ACM MM","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[He and McAuley, 2016] R. He and J. McAuley. Vbpr: Vi- sual bayesian personalized ranking from implicit feed- back. In AAAI, 2016.","title":"Vbpr: Visual bayesian personalized ranking from implicit feedback","authors":"R He, J Mcauley","year":"2016","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[He et al., 2016] R. He, C. Packer, and J. McAuley. Learn- ing compatibility across categories for heterogeneous item recommendation. In ICDM, pages 937-942, 2016.","title":"Learning compatibility across categories for heterogeneous item recommendation","authors":"R He, C Packer, J Mcauley","year":"2016","venue":"ICDM","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Hou et al., 2019] M. Hou, L. Wu, E. Chen, Z. Li, V. W. Zheng, and Q. Liu. Explainable fashion recommendation: A semantic attribute region guided approach. In IJCAI, pages 4681-4688. AAAI Press, 2019.","title":"Explainable fashion recommendation: A semantic attribute region guided approach","authors":"M Hou, L Wu, E Chen, Z Li, V W Zheng, Q Liu","year":"2019","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2019/650"},{"position":9,"raw":"[Hsiao and Grauman, 2018] Wei-Lin Hsiao and Kristen Grauman. Creating capsule wardrobes from fashion im- ages. In CVPR, pages 7161-7170, 2018.","title":"Creating capsule wardrobes from fashion images","authors":"Wei-Lin Hsiao, Kristen Grauman","year":"2018","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Ioffe and Szegedy, 2015] S. Ioffe and C. Szegedy. Batch normalization: Accelerating deep network training by re- ducing internal covariate shift. In ICML, pages 448-456, 2015.","title":"Batch normalization: Accelerating deep network training by reducing internal covariate shift","authors":"S Ioffe, C Szegedy","year":"2015","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Jin et al., 2019] B. Jin, E. Chen, H. Zhao, Z. Huang, Q. Liu, H. Zhu, and S. Yu. Promotion of answer value measure- ment with domain effects in community question answer- ing systems. IEEE TSMC: Systems, 2019.","title":"Promotion of answer value measurement with domain effects in community question answering systems","authors":"B Jin, E Chen, H Zhao, Z Huang, Q Liu, H Zhu, S Yu","year":"2019","venue":"IEEE TSMC: Systems","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Kang et al., 2017] W. Kang, C. Fang, Z. Wang, and J. McAuley. Visually-aware fashion recommendation and design with generative image models. In ICDM, pages 207-216. IEEE, 2017.","title":"Visually-aware fashion recommendation and design with generative image models","authors":"W Kang, C Fang, Z Wang, J Mcauley","year":"2017","venue":"ICDM","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Li et al., 2017] Y. Li, L. Cao, J. Zhu, and J. Luo. Mining fashion outfit composition using an end-to-end deep learn- ing approach on set data. IEEE TMM, 19(8):1946-1955, 2017.","title":"Mining fashion outfit composition using an end-to-end deep learning approach on set data","authors":"Y Li, L Cao, J Zhu, J Luo","year":"2017","venue":"IEEE TMM","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Li et al., 2018] Z. Li, H. Zhao, Q. Liu, Z. Huang, T. Mei, and E. Chen. Learning from history and present: Next- item recommendation via discriminatively exploiting user behaviors. In ACM SIGKDD, page 1734-1743, 2018.","title":"Learning from history and present: Nextitem recommendation via discriminatively exploiting user behaviors","authors":"Z Li, H Zhao, Q Liu, Z Huang, T Mei, E Chen","year":"2018","venue":"ACM SIGKDD","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Liu et al., 2010] C. Liu, L. Sharan, E. H. Adelson, and R. Rosenholtz. Exploring features in a bayesian frame- work for material recognition. In CVPR, pages 239-246. IEEE, 2010.","title":"Exploring features in a bayesian framework for material recognition","authors":"C Liu, L Sharan, E H Adelson, R Rosenholtz","year":"2010","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Liu et al., 2017] Q. Liu, S. Wu, and L. Wang. Deepstyle: Learning user preferences for visual recommendation. In SIGIR, pages 841-844. ACM, 2017.","title":"Deepstyle: Learning user preferences for visual recommendation","authors":"Q Liu, S Wu, L Wang","year":"2017","venue":"SIGIR","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Liu et al., 2018] Q. Liu, H. Zhao, L. Wu, Z. Li, and E. Chen. Illuminating recommendation by understanding the ex- plicit item relations. JCST, 33(4):739-755, 2018.","title":"Illuminating recommendation by understanding the explicit item relations","authors":"Q Liu, H Zhao, L Wu, Z Li, E Chen","year":"2018","venue":"JCST","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Lo et al., 2019] L. Lo, C. Liu, R. Lin, B. Wu, H. Shuai, and W. Cheng. Dressing for attention: Outfit based fashion popularity prediction. In ICIP, pages 3222-3226. IEEE, 2019.","title":"Dressing for attention: Outfit based fashion popularity prediction","authors":"L Lo, C Liu, R Lin, B Wu, H Shuai, W Cheng","year":"2019","venue":"ICIP","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Ma et al., 2019] J. Ma, Z. Shou, A. Zareian, H. Mansour, A. Vetro, and S. Chang. Cdsa: Cross-dimensional self- attention for multivariate, geo-tagged time series imputa- tion. arXiv preprint arXiv:1905.09904, 2019.","title":"Cdsa: Cross-dimensional selfattention for multivariate, geo-tagged time series imputation","authors":"J Ma, Z Shou, A Zareian, H Mansour, A Vetro, S Chang","year":"2019","venue":null,"doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[McAuley et al., 2015a] J. McAuley, R. Pandey, and J. Leskovec. Inferring networks of substitutable and complementary products. In ACM SIGKDD, pages 785-794. ACM, 2015.","title":"Inferring networks of substitutable and complementary products","authors":"Mcauley","year":"2015","venue":"ACM SIGKDD","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[McAuley et al., 2015b] J. McAuley, C. Targett, Q. Shi, and A Van D. H. Image-based recommendations on styles and substitutes. In SIGIR, pages 43-52. ACM, 2015.","title":"Image-based recommendations on styles and substitutes","authors":"Mcauley","year":"2015","venue":"SIGIR","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Robbins and Monro, 1951] H. Robbins and S. Monro. A stochastic approximation method. The annals of mathe- matical statistics, pages 400-407, 1951.","title":"A stochastic approximation method","authors":"H Robbins, S Monro","year":"1951","venue":"The annals of mathematical statistics","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Rudolph et al., 2016] M. Rudolph, F. Ruiz, S. Mandt, and D. Blei. Exponential family embeddings. In NIPS, pages 478-486, 2016.","title":"Exponential family embeddings","authors":"M Rudolph, F Ruiz, S Mandt, D Blei","year":"2016","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Song et al., 2017] X. Song, F. Feng, J. Liu, Z. Li, L. Nie, and J. Ma. Neurostylist: Neural compatibility model- ing for clothing matching. In ACM MM, pages 753-761. ACM, 2017.","title":"Neurostylist: Neural compatibility modeling for clothing matching","authors":"X Song, F Feng, J Liu, Z Li, L Nie, J Ma","year":"2017","venue":"ACM MM","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Szegedy et al., 2016] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna. Rethinking the inception archi- tecture for computer vision. In CVPR, pages 2818-2826, 2016.","title":"Rethinking the inception architecture for computer vision","authors":"C Szegedy, V Vanhoucke, S Ioffe, J Shlens, Z Wojna","year":"2016","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Tan et al., 2004] P. Tan, V. Kumar, and J. Srivastava. Se- lecting the right objective measure for association analy- sis. Information Systems, 29(4):293-313, 2004.","title":"Selecting the right objective measure for association analysis","authors":"P Tan, V Kumar, J Srivastava","year":"2004","venue":"Information Systems","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Uijlings et al., 2013] Jasper RR Uijlings, Koen EA Van De Sande, Theo Gevers, and Arnold WM Smeulders. Se- lective search for object recognition. IJCV, pages 154- 171, 2013.","title":"Selective search for object recognition","authors":"Uijlings","year":"2013","venue":"IJCV","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Vasileva et al., 2018] M. I. Vasileva, B. A. Plummer, K. Du- sad, S. Rajpal, R. Kumar, and D. Forsyth. Learning type- aware embeddings for fashion compatibility. In ECCV, pages 390-405, 2018.","title":"Learning typeaware embeddings for fashion compatibility","authors":"M I Vasileva, B A Plummer, K Dusad, S Rajpal, R Kumar, D Forsyth","year":"2018","venue":"ECCV","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Vaswani et al., 2017] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N Gomez, Ł. Kaiser, and I. Polo- sukhin. Attention is all you need. In NIPS, pages 5998- 6008, 2017.","title":"Attention is all you need","authors":"A Vaswani, N Shazeer, N Parmar, J Uszkoreit, L Jones, A Gomez, Ł Kaiser, I Polosukhin","year":"2017","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Veit et al., 2015] A. Veit, B. Kovacs, S. Bell, J. McAuley, K. Bala, and S. Belongie. Learning visual clothing style with heterogeneous dyadic co-occurrences. In ICCV, pages 4642-4650, 2015.","title":"Learning visual clothing style with heterogeneous dyadic co-occurrences","authors":"A Veit, B Kovacs, S Bell, J Mcauley, K Bala, S Belongie","year":"2015","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":31,"raw":"[Wang et al., 2019] X. Wang, B. Wu, and Y. Zhong. Outfit compatibility prediction and diagnosis with multi-layered comparison network. In ACM MM, pages 329-337, 2019.","title":"Outfit compatibility prediction and diagnosis with multi-layered comparison network","authors":"Wang","year":"2019","venue":"ACM MM","doi":null,"url":null,"internal_url":null},{"position":32,"raw":"[Wu et al., 2019] L. Wu, Z. Li, H. Zhao, Z. Pan, Q. Liu, and E. Chen. Estimating early fundraising performance of innovations via graph-based market environment model. arXiv preprint arXiv:1912.06767, 2019.","title":"Estimating early fundraising performance of innovations via graph-based market environment model","authors":"L Wu, Z Li, H Zhao, Z Pan, Q Liu, E Chen","year":"2019","venue":null,"doi":null,"url":null,"internal_url":null},{"position":33,"raw":"[Yu et al., 2018] W. Yu, H. Zhang, X. He, X. Chen, L. Xiong, and Z. Qin. Aesthetic-based clothing recommendation. In WWW, pages 649-658, 2018.","title":"Aesthetic-based clothing recommendation","authors":"W Yu, H Zhang, X He, X Chen, L Xiong, Z Qin","year":"2018","venue":"WWW","doi":null,"url":null,"internal_url":null},{"position":34,"raw":"[Yu et al., 2019] H. Yu, L. Litchfield, T. Kernreiter, S. Jolly, and K. Hempstalk. Complementary recommendations: A brief survey. In HPBD, pages 73-78, May 2019.","title":"Complementary recommendations: A brief survey","authors":"H Yu, L Litchfield, T Kernreiter, S Jolly, K Hempstalk","year":"2019","venue":"HPBD","doi":null,"url":null,"internal_url":null},{"position":35,"raw":"[Zhang et al., 2018] Y. Zhang, H. Lu, W. Niu, and J. Caver- lee. Quality-aware neural complementary item recommen- dation. In RecSys, pages 77-85. ACM, 2018.","title":"Quality-aware neural complementary item recommendation","authors":"Y Zhang, H Lu, W Niu, J Caverlee","year":"2018","venue":"RecSys","doi":null,"url":null,"internal_url":null},{"position":36,"raw":"[Zhao et al., 2016] L. Zhao, Z. Lu, S. J. Pan, and Q. Yang. Matrix factorization+ for movie recommendation. In IJ- CAI, pages 3945-3951, 2016.","title":"Matrix factorization+ for movie recommendation","authors":"L Zhao, Z Lu, S J Pan, Q Yang","year":"2016","venue":"IJ-CAI","doi":null,"url":null,"internal_url":"/proceedings/2016/555"},{"position":37,"raw":"[Zhao et al., 2017] T. Zhao, J. McAuley, M. Li, and I. King. Improving recommendation accuracy using networks of substitutable and complementary products. In IJCNN, pages 3649-3655, 2017.","title":"Improving recommendation accuracy using networks of substitutable and complementary products","authors":"T Zhao, J Mcauley, M Li, I King","year":"2017","venue":"IJCNN","doi":null,"url":null,"internal_url":null},{"position":38,"raw":"[Zhao et al., 2019] H. Zhao, B. Jin, Q. Liu, Y. Ge, E. Chen, X. Zhang, and T. Xu. Voice of charity: Prospecting the donation recurrence & donor retention in crowdfunding. IEEE TKDE, 2019.","title":"Voice of charity: Prospecting the donation recurrence & donor retention in crowdfunding","authors":"H Zhao, B Jin, Q Liu, Y Ge, E Chen, X Zhang, T Xu","year":"2019","venue":"IEEE TKDE","doi":null,"url":null,"internal_url":null},{"position":39,"raw":"[Zheng et al., 2009] J. Zheng, X. Wu, J. Niu, and A. Boli- var. Substitutes or complements: another step forward in recommendations. In ACM conference on Electronic com- merce, pages 139-146. ACM, 2009.","title":"Substitutes or complements: another step forward in recommendations","authors":"J Zheng, X Wu, J Niu, A Bolivar","year":"2009","venue":"ACM conference on Electronic commerce","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}