Xiaomeng Wang

Xiaomeng Wang

Auto-generated profile, based on publicly available information

Xiaomeng Wang

Flytek Co., Ltd., Hefei, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 March 2023
Verified
Conference Paper
Open access
XWXiaomeng Wang
Xiaomeng Wang
Tong XuTong Xu
Tong Xu
SWShiwei Wu
Shiwei Wu
As a widely studied topic in recommender systems, collaborative filtering (CF) methods help users discover potential items of interest by assuming that behavioral similar users would have similar preferences on items. A recent trend is to develop models based on graph neural networks (GNNs). One limitation of existing methods is that they predict user preferences only by modeling user-item interactions, failing to identify user-item relations at the finer-grained level of semantics. For more accurate and explainable recommendation, it is compulsory to take side information into account and model user-item relations at the granularity of semantics. In this paper, we investigate the utility of semantic scene graphs in video recommendation scenario, which provide detailed, graph-based annotations of social situations depicted in movie clips. We propose a new method named Scene Graph Attention Network (SGAT), which explicitly models user-item relations at the granularity of semantics by capturing certain aspects of particular interest to the user from semantic scene graphs of historical items. Empirical results on public datasets show that SGAT significantly outperforms the state-of-the-art methods like LightGCN [2] and DGCF [3]. Further studies verify the effectiveness of SGAT and the explainability for predictions.

10 March 2025 in ACM Transactions on Multimedia Computing, Communications, and Applications
Verified
Journal Article
Open access
JCJiayue Chen
Jiayue Chen
XWXiaomeng Wang
Xiaomeng Wang
Tong XuTong Xu
Tong Xu
SWShiwei Wu
Shiwei Wu
Knowledge-aware video recommendation requires the ability of associating external knowledge to capture high-order connectivities between users and videos. One limitation of existing methods is that they only extract user interests at a granular level of relational paths by modeling high-order connectivities, which are coarse-grained in user interest modeling, failing to identify user–video relations at a finer grained level of semantics. In this article, we investigate the utility of semantic scene graphs in video recommendation scenario, which provide detailed, graph-based annotations of social situations depicted in video clips. We propose a new method named Scene-Centric Multi-Level Interest Miner (SCIMiner) , which explicitly models multi-level user interests. Specifically, we construct user–video graph, knowledge graph, and semantic scene graphs as hierarchical heterogeneous graphs. On top of the hierarchical graph representation, we propose two information aggregation strategies to capture user interests from different levels. Knowledge-aware aggregation scheme extracts coarse-grained user interests by aggregating relational paths in high-order connectivities, while scene-aware aggregation scheme models fine-grained user interests by capturing clip-level semantic commonality of user favorite videos. Furthermore, we adaptively distill complementary information about multi-level user interests extracted by different aggregation schemes and encode them into the representations of users and videos. Empirical results show that SCIMiner significantly outperforms the state-of-the-art methods. Further studies verify the complementary benefits of knowledge graph with semantic scene graphs in video recommendation scenario and the finer grained explainability for predictions.