ABSTRACT As Mashup technology advances, the number of Web APIs available continues to increase year by year. As a result, identifying and selecting suitable Web APIs from the wide array of options remains a challenging issue. To address this issue, interactive Web API recommendation has been introduced to simplify the process of Web API selection, assisting users or developers in selecting Web APIs to meet their business or software development needs. Although recent graph‐based representation learning methods have advanced interactive Web API recommendation, deriving accurate representations for Mashups and Web APIs remains a challenging issue. This is largely because existing graph models struggle with the extreme sparsity of Mashup‐API interactions and the prevalent long‐tail distribution. As a result, their performance is often limited. To enhance the recommendation performance, we propose an interactive Web API recommendation approach based on long tail and composition‐supervised LightGCL, named LTCS‐LightGCL. LTCS‐LightGCL derives Mashup and Web API embeddings by utilizing a simple yet effective graph contrastive learning paradigm LightGCL in which long tail and composition‐supervised learning is designed to guide the embedding process effectively. Extensive experiments on a real‐world dataset demonstrate that LTCS‐LightGCL outperforms the baseline methods. The source code of the model can be obtained at https://github.com/IntelligentServiceLab/LTCS‐LightGCL .
Xiang et al. (Tue,) studied this question.