University of Kashan Soft Computing Journal 2322-3707 9 2 2021 02 19 A method to simplify patterns in web services compositions and select optimal probabilistic composition A method to simplify patterns in web services compositions and select optimal probabilistic composition 44 71 111554 10.22052/scj.2021.243188.1003 FA Narges Zahiri Faculty of Electrical and Computer Engineering, Department of Computer Engineering, University of Kashan, Kashan, Iran Seyed Morteza Babamir Faculty of Electrical and Computer Engineering, Department of Computer Engineering, University of Kashan, Kashan, Iran Journal Article 2021 06 02 One of the most challenging issues in web services is their composition, which is presented as a graph to show the interaction between services. Each node in such a graph is called an abstract web service with a specific function and undetermined quality attributes. For each abstract service, there is a set of candidate services with the same function but different quality attributes. Selecting a candidate web service for each abstract service leading to an optimal combination is an NP-hard problem; hence, heuristic algorithms should be used to resolve it. Several methods have been proposed to select optimal web service composition, but most of them don't support the probability structure. Among others, one method supports a probability structure that is not scalable for large graphs, is constraint based, and analyzes each path of the graph separately. This paper presents an integrated scalable multi-objective approach for analyzing graph where not only two new patterns of nested loops and parallel loops are dealt with but also performance is improved by representing a method for simplifying web-service compositions. In this method, to select optimal web services and to respect scalability, evolutionary algorithms NSGAII and SPEAII are used. In the proposed method, first in conditional graphs, each path is traversed according to its probability and then NSGAII is used to determine the best path in the graph and find better solutions. The proposed method was compared with the best known method; results showed the proposed method enjoys 30% improvement in reliability and 121 milliseconds in response time. One of the most challenging issues in web services is their composition, which is presented as a graph to show the interaction between services. Each node in such a graph is called an abstract web service with a specific function and undetermined quality attributes. For each abstract service, there is a set of candidate services with the same function but different quality attributes. Selecting a candidate web service for each abstract service leading to an optimal combination is an NP-hard problem; hence, heuristic algorithms should be used to resolve it. Several methods have been proposed to select optimal web service composition, but most of them don't support the probability structure. Among others, one method supports a probability structure that is not scalable for large graphs, is constraint based, and analyzes each path of the graph separately. This paper presents an integrated scalable multi-objective approach for analyzing graph where not only two new patterns of nested loops and parallel loops are dealt with but also performance is improved by representing a method for simplifying web-service compositions. In this method, to select optimal web services and to respect scalability, evolutionary algorithms NSGAII and SPEAII are used. In the proposed method, first in conditional graphs, each path is traversed according to its probability and then NSGAII is used to determine the best path in the graph and find better solutions. The proposed method was compared with the best known method; results showed the proposed method enjoys 30% improvement in reliability and 121 milliseconds in response time. https://scj.kashanu.ac.ir/article_111554_4f783773cf97d6566014e37d5795ced4.pdf