Modeling Multi-Purpose Sessions for Next-Item Recommendations via Mixture-Channel Purpose Routing Networks

Abstract

A session-based recommender system (SBRS) suggests the next item by modeling the dependencies between items in a session. Most of existing SBRSs assume the items inside a session are associated with one (implicit) purpose. However, this may not always be true in reality, and a session may often consist of multiple subsets of items for different purposes (e.g., breakfast and decoration). Specifically, items (e.g., bread and milk) in a subsethave strong purpose-specific dependencies whereas items (e.g., bread and vase) from different subsets have much weaker or even no dependencies due to the difference of purposes. Therefore, we propose a mixture-channel model to accommodate the multi-purpose item subsets for more precisely representing a session. Filling gaps in existing SBRSs, this model recommends more diverse items to satisfy different purposes. Accordingly, we design effective mixture-channel purpose routing networks (MCPRN) with a purpose routing network to detect the purposes of each item and assign it into the corresponding channels. Moreover, a purpose specific recurrent network is devised to model the dependencies between items within each channel for a specific purpose. The experimental results show the superiority of MCPRN over the state-of-the-art methods in terms of both recommendation accuracy and diversity.

Improving End-to-End Sequential Recommendations with Intent-aware Diversification

“…• NARM: An RNN-based model that applies an attention mechanism to capture users' main purposes from the hidden states and combines it with sequential behavior as final representations of users' current preferences [24], which shares a similar spirits as IDSR when calculating the relevance scores for items. • MCPRN: The most recently proposed method that models users' multiple purposes in a session [42]. The authors claim that they can improve the performance over the state-of-the-art methods in terms of both accuracy and diversity.…”

“…The studies mentioned above ignore the fact that users might have multiple intents reflected in their sequential behavior. Wang et al [42] have proposed a mixture-channel purpose routing networks (MCPRNs) to capture users' different intents in a given session. MCPRN first applies a purpose routing network to detect multiple purposes of a user and then models the items with a mixture-channel RNN, where each channel RNN models the item dependencies for a specific purpose.…”

Personalizing Diversity Versus Accuracy in Session-Based Recommender Systems

“…In another study [48], some regularization terms are proposed to be added to the RankALS and RankSGD approaches to generate a more diverse recommendation list. Wang et al [49] proposed the mixture-channel purpose routing networks (MCPRNs) that can cover multiple purposes of the user in the session by recommending a more diverse recommendation list. They showed that their model has better accuracy and diversity compared to the baselines.…”

When algorithm selection meets Bi-linear Learning to Rank: accuracy and inference time trade off with candidates expansion

“…Having known the partial differential of the chain rule, we can update the weight matrix W and algorithm latent matrix V by the following updating rule Eqs. (20) and (21) , where η is the learning rate.…”

“…When applying RS approaches in the AS problems, we need to note that the recorded algorithms' performances on problem instances are usually in much smaller size. Thence the state-of-the-art deep learning and transaction embedding techniques in the large-scale session-based RS [19,20] are not suitable for AS scenarios. On the contrary, shallow ML approaches from RS are more adaptable.…”