Sequence-Aware Service Recommendation Based on Graph Convolutional Networks

With the increasing number of Web services, how to quickly and efficiently recommend services that meet user needs has become a major research hotspot in the field of service computing. However, most existing service recommendation methods do not sufficiently account for differences in the importance of service information and often ignore the dynamic changes in user development requirements. To address these issues, this paper proposes a sequence-aware service recommendation method based on graph convolutional network (GCN-SSR). The method first utilizes the two-tower model to mine the similarity of service features and construct a service network graph. In the graph convolution process, a query-aware self-attention mechanism is proposed to aggregate service information and avoid the problem of compound embedding space. Further, we adopt the graph pooling method of coarsening strategy to extract service information. Moreover, the service itself and its location information are combined to capture focus services change. Finally, we conduct experiments to evaluate the proposed method and demonstrate the effectiveness of the method on a real dataset.