Web Service Recommendation via Combining Doc2Vec-Based Functionality Clustering and DeepFM-Based Score Prediction

Due to the rapid growth in both the number and diversity of Web services on the Internet, it becomes increasingly difficult for developer to find the desired and appropriate Web services for Mashup creation. Even if the existing approaches show improvements in Web APIs recommendation, it is still challenging to recommend Web APIs with high accuracy and good diversity. Some of them integrate functionality clustering and the quality of service to recommend Web APIs for Mashup creation, but do not consider the high-order composition interaction relationship among functionality information, quality attributes. In this paper, we propose a novel Web APIs recommendation method via integrating the functionality clustering of service and the quality of service. In this method, it firstly obtains the functionality clustering by using Doc2Vec to cluster the description document of Web APIs. Then, the deep factorization machine model is used to extract the multi-dimension quality attributes of service and mine the high-order composition interaction relationship between them. Finally, the comparative experiments are performed on ProgrammableWeb dataset and experimental results show that our method significantly improves the performance of Web API recommendation in term of precision, recall, purity, entropy, DCG and HMD.