With the rapid development of service computing technology, how to recommend the desirable Web APIs to developers from the large number of APIs is a challenge. The traditional methods based on collaborative filtering are limited by data sparsity. With the support of multi-dimensional relational feature modeling, graph neural network-based methods are proposed to mitigate data sparsity, but their convolution nature may amplify the noise effect. Therefore, how to simultaneously reduce the impact of sparsity and noisy data has been an open question in the field of API recommendation. To address the problem, this paper proposes a self-supervised learning method based on multi-view hypergraph for Web API recommendation. First, the Mashup-API interaction graph is transformed into a hypergraph, and the hyperedges are used as intermediate hubs to transfer messages between nodes, maintaining the global collaboration effect between Mashup and API nodes. Then, a multi-view strategy is adopted to generate embeddings of Mashups and APIs through information fusion, by which recommendation probability is derived by dot product between the embeddings of Mashups and Web APIs. To train the model parameters effectively, a self-supervised learning method is used to reduce the effect of noisy data to improve the embeddings. Extensive experiments are conducted on a real-world dataset, and the experimental results show that the proposed model outperforms the baselines.

With the continually increasing number of Web services, it becomes a challenging task to efficiently and accurately provide Web services that meet developers' functional requirements. Existing heterogeneous graph-based service recommendation methods simply utilize the heterogeneous structural features of the service network and suffer from the missing and blurring of service interaction semantic information due to the characteristics of meta-paths. In fact, service node description documents contain fine-grained semantics generated by multifaceted topic-aware factors, but few efforts are committed to mining them. Therefore, a Web service recommendation method via combining topic-aware heterogeneous graph representation and interactive semantic enhancement is proposed in this paper. It employs an alternating two-step aggregation mechanism, including meta-path instance intra-decomposition and meta-path inter-integration, which uniquely aggregates topic-aware factors according to the inferred topic distributions while preserving structural semantics. Additionally, it introduces the topic prior knowledge guidance module to improve the quality of the inference's topic factors. Simultaneously, the method designs the interactive semantic enhancement module to address the missing and blurring of service interaction semantic information caused by meta-paths. The module explores complex interaction patterns among services and utilizes personalized knowledge meta-network to enhance contrastive learning of service interaction semantics, allowing the personalized knowledge transformer with adaptive contrastive enhancement. The experimental results on the real dataset of ProgrammableWeb show that compared with the other nine methods, the proposed method has better service recommendation performance on evaluation metrics HR and NDCG representing accuracy and satisfaction, respectively.