TRec: Sequential Recommender Based On Latent Item Trend Information

Abstract

Recommendation system plays an important role in online web applications. Sequential recommender further models user short-term preference through exploiting information from latest user-item interaction history. Most of the sequential recommendation methods neglect the importance of ever-changing item popularity. We propose the model from the intuition that items with most user interactions may be popular in the past but could go out of fashion in recent days. To this end, this paper proposes a novel sequential recommendation approach dubbed TRec, TRec learns item trend information from implicit user interaction history and incorporates item trend information into next item recommendation tasks. Then a self-attention mechanism is used to learn better node representation. Our model is trained via pairwise rank-based optimization. We conduct extensive experiments with seven baseline methods on four benchmark datasets, The empirical result shows our approach outperforms other stateof-the-art methods while maintains a superiorly low runtime cost. Our study demonstrates the importance of item trend information in recommendation system designs, and our method also possesses great efficiency which enables it to be practical in real-world scenarios.

Item trend learning for sequential recommendation system using gated graph neural network

“…In the next section, we will discuss three schemes to implement the function F tr ðÁÞ. The first implementation of F tr ðÁÞ is first introduced in our previous paper [10] which TRec is the first time proposed. This scheme could be viewed as a simplified model of the second and the third implementation of F tr ðÁÞ which are going to be elaborated in the forthcoming sections.…”

“…Aiming to deal with the aforementioned problems, we propose ways to model the item trend information; different approaches are used to achieve this goal. In addition to the one we proposed previously [10], different approaches for the item trend representation modeling with gated graph neural network are discussed as well. Graph neural network, as a powerful tool for the node embeddings learning, is utilized to generate the reliable and accurate item trend representation for candidate items; this technique is capable of providing rich local contextual information by encoding node features, while vanilla GNN is best for non-sequential input, for example, features like height, weight and gender of a person; there's no sequential relation among these features.…”