P-Companion

Abstract

Complementary product recommendation (CPR), aiming at providing product suggestions that are often bought together to serve a joint demand, forms a pivotal component of e-commerce service, however, existing methods are far from optimal. Given one product, how to recommend its complementary products of different types is the key problem we tackle in this work. We first conduct an analysis to correct the inaccurate assumptions adopted by existing work to show that co-purchased products are not always complementary and further propose a new strategy to generate clean distant supervision labels for CPR modeling. Moreover, to bridge in the gap from existing work that CPR does not only need relevance modeling but also requires diversity to fulfill the whole purchase demand, we develop a deep learning framework, P-Companion, to explicitly model both relevance and diversity. More specifically, given one product with its product type, P-Companion first uses an encoderdecoder network to predict multiple complementary product types, and then a transfer metric learning network is developed to project the embedding of query product to each predicted complementary product type subspace and further learn the complementary relationship based on the distant supervision labels. The whole framework can be trained from end-to-end and is robust to coldstart products attributed to a novel pretrained product embedding module named Product2vec, based on graph attention networks. Extensive offline experiments show that P-Companion outperforms state-of-the-art baselines by 7.1% increase on the Hit@10 score with well-controlled diversity. Production-wise, we deploy P-Companion to provide online recommendations for over 200M products at Amazon and observe significant gains on product sales and profit. CCS CONCEPTS • Information systems → Recommender systems; Online advertising; Online shopping; • Computing methodologies → Knowledge representation and reasoning.

Desire for Birth Companionship Among Pregnant Women Attending Antenatal Care in Debremarkos City, Northwest Ethiopia: Magnitude and Associated Factors

“…The possible explanation might be due to women with no bad obstetric history mostly cautious about their pregnancy and highly adhered to the maternity continuum of care ( 48 ). Birth companionship is one of the recommended interventions by the WHO and sermonized by health care providers during the antenatal care period ( 8 ). Therefore, the more adhered to antenatal care visits the more positive impression about birth companionship and become the more desired to have during labor and delivery.…”

“…Emotional and social support of women's choice is core to the experience of care and to achieve positive person-centered health outcomes ( 7 ). Thus, World Health Organization (WHO) recommends that every woman is offered the option to experience labor and childbirth with a companion of her choice ( 8 ).…”

Personalized diversification of complementary recommendations with user preference in online grocery

“…In addition to the co-purchase data, many types of auxiliary data are incorporated into the modeling, such as the multimodal data of items (Zhang et al, 2018 ) and the shopping context (Xu et al, 2019 ). Diversified complementary recommendation is studied in Hao et al ( 2020 ) by leveraging the product-type information to improve the diversity. However, it focuses on the diversified recall process rather than the ranking process as our article targets.…”

“…Many complementary item recommendation models mainly focused on learning the complementarity between items rather than the personalized adjustment of diversity of complementary item recommendations (McAuley et al, 2015 ; Barkan and Koenigstein, 2016 ; Wan et al, 2018 ; Wang et al, 2018 ; Zhang et al, 2018 ; Xu et al, 2019 ; Liu et al, 2020 ). Diversification of complementary item recommendations has been recently addressed in Hao et al ( 2020 ) by considering the item type and categories. Unfortunately, it cannot distinguish the demand of users' shopping intent by surfacing more heterogeneous complementary items for exploratory shopping intent or more homogeneous complementary items for conventional shopping intent.…”

Towards High-Order Complementary Recommendation via Logical Reasoning Network

“…Complementary recommendation mainly relies on learning the underlying relationship among products, which is often characterized by the co-purchase and co-view patterns from customer engagement traffic. To extract the complementary relationship between products, there exists a few prior works such as association mining [1], item-based collaborative filtering [2], representation learning [3][4][5][6]. However, the traditional data-mining and collaborative filtering based approaches [1,2] can only learn the symmetrical co-purchase relationship between products; most existing deep models rely on auxiliary information (e.g., scene background features required in [3], user purchase sequence information required in [4], product catalog information [5]) which limits their Fig.…”