POG: Personalized Outfit Generation for Fashion Recommendation at Alibaba iFashion

Chen, Wen; Huang, Pipei; Xu, Jiaming; Guo, Xin; Guo, Cheng; Sun, Fei; Li, Chao; Pfadler, Andreas; Zhao, Huan; Zhao, Binqiang

doi:10.1145/3292500.3330652

Cited by 148 publications

(76 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(1) We use the Amazon-Book and Last-FM datasets released by KGAT [41]; And (2) we further introduce the Alibaba-iFashion dataset [8] to investigate the effectiveness of item knowledge. This is a fashion outfit dataset collected from Alibaba online shopping systems.…”

Section: Experimental Settingsmentioning

confidence: 99%

Learning Intents behind Interactions with Knowledge Graph for Recommendation

Wang

Huang

Wang

et al. 2021

Proceedings of the Web Conference 2021

332

119

View full text Add to dashboard Cite

Knowledge graph (KG) plays an increasingly important role in recommender systems. A recent technical trend is to develop endto-end models founded on graph neural networks (GNNs). However, existing GNN-based models are coarse-grained in relational modeling, failing to (1) identify user-item relation at a fine-grained level of intents, and (2) exploit relation dependencies to preserve the semantics of long-range connectivity.In this study, we explore intents behind a user-item interaction by using auxiliary item knowledge, and propose a new model, Knowledge Graph-based Intent Network (KGIN). Technically, we model each intent as an attentive combination of KG relations, encouraging the independence of different intents for better model capability and interpretability. Furthermore, we devise a new information aggregation scheme for GNN, which recursively integrates the relation sequences of long-range connectivity (i.e., relational paths). This scheme allows us to distill useful information about user intents and encode them into the representations of users and items. Experimental results on three benchmark datasets show that, KGIN achieves significant improvements over the state-ofthe-art methods like KGAT [41], and CKAN [47]. Further analyses show that KGIN offers interpretable explanations for predictions by identifying influential intents and relational paths. The implementations are available at https://github.com/ huangtinglin/Knowledge_Graph_based_Intent_Network.

show abstract

Section: Experimental Settingsmentioning

confidence: 99%

Learning Intents behind Interactions with Knowledge Graph for Recommendation

Wang

Huang

Wang

et al. 2021

Proceedings of the Web Conference 2021

332

119

View full text Add to dashboard Cite

show abstract

“…During the last decade, Recommender Systems (RSs) have been the most popular application in industry, and in the past five years, deep learning based methods have been widely used in industrial RSs, e.g., Google [2,3] and Airbnb [5]. In Alibaba, the largest ecommerce platform in China, RSs have been the key engine for its Gross Merchandise Volume (GMV) and revenues, and various deep learning based recommendation methods have been deployed in rich e-commerce scenarios [1,8,10,11,14,15,17,18]. As introduced in [15], the RSs in Alibaba are a two-stage pipeline: match and rank.…”

Section: Introductionmentioning

confidence: 99%

Behavior sequence transformer for e-commerce recommendation in Alibaba

Chen

Zhao

et al. 2019

Proceedings of the 1st International Workshop on Deep Learning Practice for High-Dimensional Sparse Data

Self Cite

304

145

View full text Add to dashboard Cite

Deep learning based methods have been widely used in industrial recommendation systems (RSs). Previous works adopt an Em-bedding&MLP paradigm: raw features are embedded into lowdimensional vectors, which are then fed on to MLP for final recommendations. However, most of these works just concatenate different features, ignoring the sequential nature of users' behaviors. In this paper, we propose to use the powerful Transformer model to capture the sequential signals underlying users' behavior sequences for recommendation in Alibaba. Experimental results demonstrate the superiority of the proposed model, which is then deployed online at Taobao and obtain significant improvements in online Click-Through-Rate (CTR) comparing to two baselines.

show abstract

“…Here we present a modified approach that handles product-to-product recommendation. Many recommendation systems, such as Alibaba's iFashion [4], use past user engagement to train complementary models [21] [9]. We attempt the challenging task of bootstrapping such a system without explicit user data.…”

Section: Recommender Systemsmentioning

confidence: 99%