Диссертации, представленные на защиту и подготовленные в НИУ ВШЭ

This thesis is devoted to context-based recommender systems, particularly to novel methods for enhancing existing algorithms and proposing new approaches for incorporating auxiliary context information, such as the concrete time of user-item interactions and its various derivatives, including time intervals between events. The problems of the next item and the next basket prediction are considered and four new methods are proposed (time-aware and time-dependent TIFU-KNN, Time-Aware Item Weighting (TAIW), time-aware GRU4Rec and TiSASRec). For knowledge-based context-aware recommender systems, a fusion of neural networks and a knowledge graph based approach,  TimeKGATLstm, is proposed. All the approaches underwent comprehensive experimental validation using state-of-the-art benchmarking datasets against competing methods, demonstrating the superiority of the proposed methods (in most cases) in terms of relevant quality metrics for recommender systems, which validates that the proposed methods of context incorporation reliably improve performance.

Nowadays, recommender systems are essential components of various consumer services, from e-commerce to social media. They help navigate through a large volume of items empowering user experience. Methods vary from classic matrix completion techniques to modern sequence models inspired by natural language processing. One of the prominent approaches is to consider the recommender system as a link prediction problem on a bipartite user-item interaction graph.The dissertation studies the adaptation of network embedding techniques to recommender systems. In the dissertation, we investigated different aspects of graph machine learning techniques, and their effect on downstream link prediction problem. We proposed an efficient strategy to incorporate node and edge features with structural information to solve the link prediction (recommendation) problem. Also, we developed novel models to preserve temporality and enable graph-based exploration for recommender systems. Proposed approaches were compared with the state-of-the-art open-source benchmarks and showed the efficiency of our approach.