Textual and Graph Representation Learning-based Service Recommendation for Application Construction

As a new computing paradigm, service-oriented computing has seen fast development in recent years. As a result, a proliferation of both traditional Web-based and cloud-based service models have appeared online. These open services will play a crucial role in developing applications. However, the difficulty of service selection in developing a service-oriented system grows in proportion to the growing number of available services. This research suggests expanding upon the foundation of the original collaborative filtering recommendation algorithm by developing a model for service recommendations based on the study of representation learning of knowledge graphs and texts. In this research, we focus on modeling the semantic features of applications and services through the use of Translating Embedding, a knowledge graph feature learning approach, and the textual features learned from topic models. We conducted experiments on a real-world dataset, and the results demonstrate that the suggested method is able to locate services that are pertinent to developer needs.