Our research reveals that multi-interest recommendation models in the matching stage tend to exhibit an under-clustered item embedding space, which leads to a low discernibility between items and hampers item retrieval. This highlights the necessity for item embedding enhancement. However, item attributes, which serve as effective side information for enhancement, are either unavailable or incomplete in many public datasets due to the labor-intensive nature of manual annotation tasks. This dilemma raises two meaningful questions: 1. Can we bypass manual annotation and directly simulate complete attribute information from the interaction data? And 2. If feasible, how can we simulate attributes with high accuracy and low complexity in the matching stage? In this paper, we first establish a theoretical feasibility that the item-attribute correlation matrix can be approximated through elementary transformations on the item co-occurrence matrix. Then, based on further formula derivation, we propose a simple yet effective module, SimEmb (Item Embedding Enhancement via Simulated Attribute), in the multi-interest recommendation of the matching stage. By simulating attributes with the co-occurrence matrix, SimEmb discards the ID-based item embedding and employs the attribute-weighted summation for item embedding enhancement. Comprehensive experiments on four benchmark datasets demonstrate that our approach notably enhances the clustering of item embedding and significantly outperforms SOTA models with an average improvement of 25.59% on Recall@20.
Multi-interest recommendation methods extract multiple interest vectors to represent the user comprehensively. Despite their success in the matching stage, previous works overlook the long-tail problem. This results in the model excelling at suggesting head items, while the performance for tail items, which make up more than 70% of all items, remains suboptimal. Hence, enhancing the tail item recommendation capability holds great potential for improving the performance of the multi-interest model. Through experimental analysis, we reveal that the insufficient context for embedding learning is the reason behind the underperformance of tail items. Meanwhile, we face two challenges in addressing this issue: the absence of supplementary item features and the need to maintain head item performance. To tackle these challenges, we propose a CoLT module (Co-occurrence embedding enhancement for Long-Tail problem) that replaces the embedding layer of existing multi-interest frameworks. By linking co-occurring items to establish "assistance relationships", CoLT aggregates information from relevant head items into tail item embeddings and enables joint gradient updates. Experiments on three datasets show our method outperforms SOTA models by 21.86% Recall@50 and improves the Recall@50 of tail items by 14.62% on average.
Existing recommendation methods based on multi-interest frameworks effectively model users from multiple aspects to represent complex user interests. However, more research still needs to be done on the behavior of users shopping for others. We propose a Multi-Demander Recommendation (MDR) model to learn different people’s interests from a sequence of actions. We first decouple the feature embeddings of items to learn the static preferences of different demanders. Next, a weighted directed global graph is constructed to model the associations among item categories. We partition short sequences by time intervals and look up category embeddings from the graph to capture dynamic intents. Finally, preferences and intentions are combined with learning the interests of different demanders. The conducted experiments demonstrate that our model improves the accuracy of recommendations.