Search

Session-based recommender systems (SBRSs) predict users' next interacted items based on their historical activities. While most SBRSs capture purchasing intentions locally within each session, capturing items' global information across different sessions is crucial in characterizing their general properties. Previous works capture this cross-session information by constructing graphs and incorporating neighbor information. However, this incorporation cannot vary adaptively according to the unique intention of each session, and the constructed graphs consist of only one type of user-item interaction. To address these limitations, we propose knowledge graph-based session recommendation with session-adaptive propagation. Specifically, we build a knowledge graph by connecting items with multi-typed edges to characterize various user-item interactions. Then, we adaptively aggregate items' neighbor information considering user intention within the learned session. Experimental results demonstrate that equipping our constructed knowledge graph and session-adaptive propagation enhances session recommendation backbones by 10%-20%. Moreover, we provide an industrial case study showing our proposed framework achieves 2% performance boost over an existing well-deployed model at The Home Depot e-platform.

Session-based recommendation has gained increasing attention in recent years, with its aim to offer tailored suggestions based on users' historical behaviors within sessions. To advance this field, a variety of methods have been developed, with ID-based approaches typically demonstrating promising performance. However, these methods often face challenges with long-tail items and overlook other rich forms of information, notably valuable textual semantic information. To integrate text information, various methods have been introduced, mostly following a naive fusion framework. Surprisingly, we observe that fusing these two modalities does not consistently outperform the best single modality by following the naive fusion framework. Further investigation reveals an potential imbalance issue in naive fusion, where the ID dominates and text modality is undertrained. This suggests that the unexpected observation may stem from naive fusion's failure to effectively balance the two modalities, often over-relying on the stronger ID modality. This insight suggests that naive fusion might not be as effective in combining ID and text as previously expected. To address this, we propose a novel alternative training strategy AlterRec. It separates the training of ID and text, thereby avoiding the imbalance issue seen in naive fusion. Additionally, AlterRec designs a novel strategy to facilitate the interaction between the two modalities, enabling them to mutually learn from each other and integrate the text more effectively. Comprehensive experiments demonstrate the effectiveness of AlterRec in session-based recommendation. The implementation is available at https://github.com/Juanhui28/AlterRec.

Popularity of social networks has rapidly increased over the past few years, and daily lives interrupt without their proper functioning. Social networking platform provide multiple interaction types between individuals, such as creating and joining groups, sending and receiving messages, sharing interests and creating friendship relationships. This paper addresses an important problem in social networks analysis and mining that is how to predict link initiation feedback in two-way networks. Relationships between two individuals in a two-way network include a link invitation from one of the individuals, which will be an established link if it is accepted by the invitee. We consider a sport gaming social networking platform and construct a multilayer social network between a number of users. The network formed by the link initiation process is on one of the layers, while the other two layers include a messaging relationships and interactions between the users. We propose a methodology to solve the link initiation feedback prediction problem in this multilayer fashion. The proposed method is based on features extracted from meta-paths, i.e. paths defined between different individuals from multiples layers in multilayer networks. We proposed a cluster-based approach to handle the sparsity issue in the dataset. Experimental results show that the proposed method can provide accurate prediction that outperforms state-of-the-art methods.

One of the major issues in signed networks is to use network structure to predict the missing sign of an edge. In this paper, we introduce a novel probabilistic approach for the sign prediction problem. The main characteristic of the proposed models is their ability to adapt to the sparsity level of an input network. The sparsity of networks is one of the major reasons for the poor performance of many link prediction algorithms, in general, and sign prediction algorithms, in particular. Building a model that has an ability to adapt to the sparsity of the data has not yet been considered in the previous related works. We suggest that there exists a dilemma between local and global structures and attempt to build sparsity adaptive models by resolving this dilemma. To this end, we propose probabilistic prediction models based on local and global structures and integrate them based on the concept of smoothing. The model relies more on the global structures when the sparsity increases, whereas it gives more weights to the information obtained from local structures for low levels of the sparsity. The proposed model is assessed on three real-world signed networks, and the experiments reveal its consistent superiority over the state of the art methods. As compared to the previous methods, the proposed model not only better handles the sparsity problem, but also has lower computational complexity and can be updated using real-time data streams.

Recommendation systems have wide-spread applications in both academia and industry. Traditionally, performance of recommendation systems has been measured by their precision. By introducing novelty and diversity as key qualities in recommender systems, recently increasing attention has been focused on this topic. Precision and novelty of recommendation are not in the same direction, and practical systems should make a trade-off between these two quantities. Thus, it is an important feature of a recommender system to make it possible to adjust diversity and accuracy of the recommendations by tuning the model. In this paper, we introduce a probabilistic structure to resolve the diversity-accuracy dilemma in recommender systems. We propose a hybrid model with adjustable level of diversity and precision such that one can perform this by tuning a single parameter. The proposed recommendation model consists of two models: one for maximization of the accuracy and the other one for specification of the recommendation list to tastes of users. Our experiments on two real datasets show the functionality of the model in resolving accuracy-diversity dilemma and outperformance of the model over other classic models. The proposed method could be extensively applied to real commercial systems due to its low computational complexity and significant performance., Comment: 19 pages, 5 figures