A5102820471- Recommender Systems and Techniques
- Advanced Graph Neural Networks
- Advanced Bandit Algorithms Research
- Topic Modeling
- Image Retrieval and Classification Techniques
- Mental Health via Writing
Shenzhen University
2023-2024
Although knowledge graph has shown their effectiveness in mitigating data sparsity many recommendation tasks, they remain underutilized context-aware recommender systems (CARS) with the specific challenges associated contextual features, i.e., feature and interaction sparsity. To bridge this gap, paper, we propose a novel pairwise intent embedding learning (PING) framework to efficiently integrate graphs into CARS. Specifically, our PING contains three modules: 1) construction module is used...
Sequential recommendation has been widely used to predict users' potential preferences by learning their dynamic user interests, for which most previous methods focus on capturing item-level dependencies. Despite the great success, they often overlook stage-level interest In real-world scenarios, interests tend be staged, e.g., following an item purchase, a user's may undergo transition into subsequent phase. And there are intricate dependencies across different stages. Meanwhile, behaviors...
In real recommendation scenarios, users often have different types of behaviors, such as clicking and buying. Existing research methods show that it is possible to capture the heterogeneous interests through behaviors. However, most multi-behavior approaches limitations in learning relationship between this paper, we propose a novel multilayer perceptron (MLP)-based sequential method, namely behavior-aware (BMLP). Specifically, has two main modules, including interest perception (HIP)...