An efficient algorithm for web service selection based on local selection in large scale

Abstract

Nowadays, services are published on the Internet is growing explosively, it causes that it is difficult to select the web services which have the same function to satisfy the users' requirements. (Quality of Service) QoS mainly includes cost, response time, availability, etc. It is also knew to be the most important non-functional standard in service selection. So how to select a best Web service on real-time for large services, it will become very important. Many researchers have adopted various heuristic and meta-heuristic algorithm to solve the problem of web service selection based on QoS. One of the algorithms is the genetic algorithm. But genetic algorithm doesn't have a good adaptability, that means when the number of services which are expected to mutation is so many, it often needs more search space. So in this paper, I will introduce the label method to select “good” services, then by genetic algorithm to get the best services selection.