Locality Sensitive Hashing Based Service Classification

Technology of Web Service Classification can be widely used in the area of service applications. With the rapid growth of services, service classification turn to work in a poor performance. In this paper, we propose an efficient algorithm based on Locality Sensitive Hashing (LSH) to solve the service classification problem. This method is capable of high dimensional service classification as well. It is introduced that how to convert WSDL to vectors based on VSM. Then services are classified using LSH. We evaluate our approach and KNN method through experiment on different datasets. Two standards precision and recall are adopted to check the quality of service classification. The experiments have shown that our algorithm has a high performance and works stably.