Application of Ant Algorithm for Software Optimization

Abstract

Article HistoryKeywords Software Optimization Ant algorithm Soft computing Technology Method. Different technologies, methods and algorithms are used when developing high-quality software systems. There are various ways to create optimal software. One of them is an ant algorithm. The ant algorithm can be attributed to the field of biomimetic. Ant algorithm is one of the most effective polynomial algorithms for finding approximate solutions, and it also solves similar route search problems in graphs. Various methods, including the ant algorithm, are used to improve software efficiency and optimize it. The essence of the concept is to analyze, use, and meta-heuristically optimize the behavior of ants in search of paths to the food source of the colony. This article analyzes the studies in this area, and explains the idea of the algorithm used, compares different ant systems, shows program code operators as ants, and applies the ant algorithm to them. As a result of applying the algorithm, the shortest path to some operators (cycle, condition) contained in the program code is found. The experiments perform good results.Contribution/Originality: The ant algorithm, various methods including the ant algorithm was studied.Analyzes the studies in this area, and explains the idea of the algorithm used, compares different ant systems, shows program code operators as ants, and applies the ant algorithm to them. Figure-1. Software efficiency characteristics. Note that biomimetic is an imitation model of the systems and elements in the nature to solve complex human problems. Living organisms have well-adapted structures and materials for natural selection and have developed over many years. The study of biomimetic technologies and their application in various fields can play an important role in the successful development of the economy.Ant algorithm (Ant Colony Optimization, ACO) is one of the most effective polynomial algorithms for finding approximate solutions, and it also solves similar route search problems in graphs. The essence of the concept is to 13 In this step, the right transition from one operator to others has to be found. As seen from the values, the probability value varies within 0, 1. P11 -[0, 0.3] P12 -[0, 0.2] P13 -[0, 0.1] P14 -[0, 0.1] P15 -[0, 0.1] P16 -[0, 0.1] P17 -[0, 0.3] Step 2. All transitions of the 2nd node are assessed. P23 -[0, 0.2] P24 -[0, 0.3] P25 -[0, 0.2] P26 -[0, 0.2] P27 -[0, 0.2] P28 -[0, 0.3]As seen from these values, their probability value also varies within 0.1.Step 3. All transitions of the 3rd node is assessed. P34 -[0, 0.5] P35 -[0, 0.2] P38 -[0, 0.1] P37 -[0, 0.2] P36 -[0, 0.1] Step 4. All transitions of the 4th node are assessed. P45 -[0, 0,5] P48 -[0, 0.3] P47 -[0, 0.2] P46 -[0, 0.2] Review of Information Engineering and Applications, 2020, 9(1): 6-17 15 Step 5. All transitions of the 5th node is assessed. P55 -[0, 0,7] P58 -[0, 0.2] P57 -[0, 0.3] Step 6. All transitions of the 6nd node are assessed. P67 -[0, 0,4] P68 -[0, 0.7]Step 7. The 7th node will have one value.P78 -[0...