The Study of Web Service Discovery: A Clustering and Differential Evolution Algorithm Approach
Zhao Hui Huang, Wei Zhao · 2019
With the rapid development of the Internet, the number of web services has a significant increase, providing various types of online services available today. However, finding the target web service in a quick and accurate method to meet the needs of users is becoming the major problem of the service discovery. In this study, we proposed a new method of web service discovery, which is based on the clustering and differential evolution algorithm. To carry out our study, firstly, we clustered the service set and identified the most similar class to narrow down the searching scope by calculating the semantic similarity between each class and the query. Second, the similarities of each service in the target class to the query are calculated by using different measurement methods, including Cosine similarity, Euclidean distance, Manhattan distance, Chebyshev distance and Bray Curtis distance. These results are used as the input of the differential evolution algorithm for our iterative simulation. Our findings show that compared with the traditional methods, our proposed approach that integrates the benefits of the clustering and differential evolution algorithm has a lower error, and is more consistent with the users' requirements. More importantly, relevant web services can be retrieved more efficiently and accurately.