Functional and Structural Fusion based Web API Recommendations in Heterogeneous Networks
Xuanye Wang, Meng Xi, Jianwei Yin
Abstract
With the increasing development of cloud computing, a large number of innovative Mashup applications and Web APIs have emerged on the Internet. The expansion of technology and information presents a significant challenge to the discovery of Web APIs from multiple service ecosystems. Various Web API recommendation methods have been proposed in mashup creation, but most either assign equal weight to model factorization interactions or solely rely on requirements information for API recommendation. Unfortunately, these methods face several challenges, such as explicit and implicit dependencies among APIs, ambiguous API semantics, and the undervaluation of tail APIs. In this work, we propose a Functional and Structural Fusion Model (FSFM) based on Web API recommendation for Mashup creation. To address the above deficiencies, we first design the structural interaction component to encode the latent structural relationships in the heterogeneous network of Mashups and APIs and capture the topological structure signals between different Mashups and APIs. Then, we introduce the functional semantic component to generate text embedding vectors for Mashups and APIs, enhancing their requirement semantics at multiple levels of abstraction. Finally, we fuse the output vectors to obtain the list of candidate Web APIs. Experiences are performed on real datasets, and statistical results show that FSFM outperforms other state-of-the-art models in both overall and long-tail Web API recommendations.