Dual Attentive Sequential Learning for Cross-Domain Click-Through Rate Prediction

Abstract

Cross domain recommender system constitutes a powerful method to tackle the cold-start and sparsity problem by aggregating and transferring user preferences across multiple category domains. Therefore, it has great potential to improve click-through-rate prediction performance in online commerce platforms having many domains of products. While several cross domain sequential recommendation models have been proposed to leverage information from a source domain to improve CTR predictions in a target domain, they did not take into account bidirectional latent relations of user preferences across source-target domain pairs. As such, they cannot provide enhanced cross-domain CTR predictions for both domains simultaneously. In this paper, we propose a novel approach to cross-domain sequential recommendations based on the dual learning mechanism that simultaneously transfers information between two related domains in an iterative manner until the learning process stabilizes. In particular, the proposed Dual Attentive Sequential Learning (DASL) model consists of two novel components Dual Embedding and Dual Attention, which jointly establish the two-stage learning process: we first construct dual latent embeddings that extract user preferences in both domains simultaneously, and subsequently provide cross-domain recommendations by matching the extracted latent embeddings with candidate items through dual-attention learning mechanism. We conduct extensive offline experiments on three real-world datasets to demonstrate the superiority of our proposed model, which significantly and consistently outperforms several state-of-the-art baselines across all experimental settings.

Recommendation Systems: An Insight Into Current Development and Future Research Challenges

“…Base Cat. Enhancements HRNN-meta [157] GRU, RNN CE, CA Hierarchical sessions for inter-and intra-session relations CTA [158] bi-RNN, ATT CA Modeling temporal dynamics with contextualized self-attention CSRM [159] RNN, ATT CF (I) Collaborative information in session-based recommendation DASL [160] GRU, ATT CA Cross-domain data recommendation with dual attention CAML [161] GRU, ATT, FM CE Enhance accuracy and explainability through cross-task learning CatDM [162] LSTM, ATT CE, CA Integration of POI categories, geographical and temporal influences ESRM-KG [163] Tran, GRU CE, CA Multi-task learning by generating keywords to extract intent SDM [164] LSTM, ATT CE, CA Dynamic preference modeling using short-and long-term behaviors SSRM [165] GRU most popularly implementing a gating mechanism such as Long Short-Term Memory (LSTM) units [170] and Gated Recurrent Units (GRU) [171], such as to solve the various challenges faced by vanilla RNNs (e.g., the vanishing gradient problem). Furthermore, recent years have seen a dramatic increase in the proposal of approaches utilizing the attention mechanism [172], [173], which was indeed first popularized in the context of recurrent networks.…”

“…3) Other RNN-based approaches Lastly, we introduce some RNN-based methods which explore different research directions. The work by [160], for instance, explores cross-domain sequential recommendations to improve CTR accuracy. The proposed Dual Attentive Sequential Learning (DASL) learns cross-domain user representations using a dual embedding strategy, which extracts latent embeddings in both domains simultaneously through metric learning.…”

Differential Private Knowledge Transfer for Privacy-Preserving Cross-Domain Recommendation

“…Cross Domain Recommendation (CDR) is proposed to handle the cold-start and data sparsity problem commonly existed in the traditional single domain recommender systems [35]. The basic assumption of CDR is that different behavioral patterns from multiple domains jointly characterize the way users interact with items [21,34]. Thus, CDR models can improve the recommendation performance of the single domain recommender systems by transferring the knowledge learnt from other domains using auxiliary information.…”

DDGHM: Dual Dynamic Graph with Hybrid Metric Training for Cross-Domain Sequential Recommendation

“…Second, although several representative CDR approaches [2,9,15,47] address the data sparsity and cold-start problem by transferring knowledge within domains, they cannot take full advantage of sequential patterns. Third, existing CDSR methods [1,4,14,18,30] reach the breakthrough of exploring sequential dependencies and modeling structure information that bridges two domains. However, these CDSR models cannot extract and incorporate both the intra-domain and the inter-domain item transitions in a dynamical and synchronous way.…”