Performance Comparision of Artificial Intelligence Techniques in Software Reusability

Abstract

Fuzzy logic has proved its mettle in last few decades and has been used in various applications to improve the performance and embeds some intelligence into the system. Fuzzy logic also solves the problem of non linear systems and handles them with great efficiency and provides robustness to the system. However our aim always lies in achieving the improved solution and there are different hybrid algorithms.. In this thesis the recent data based artificially intelligent techniques like fuzzy and Neurofuzzy have been customized and used .The application/case study has been taken from a research paper which appeared in a reputed general. The case study deals with reusability of software components. The attributes are coupling, volume, complexity, regularity and reuse frequency. In such data search application the design and developed neuro-fuzzy hybrid algorithm has shown its superiority because it includes the advantages of fuzzy as well as neural networks. Neuro -fuzzy algorithms is definitely superior to fuzzy algorithm as it inherits adaptability and learning. From the simulation and the result obtained in this thesis .it has been shown that the percentage average error is less in neuro-fuzzy model. Neuro-fuzzy algorithm has yielded accuracy greater than the accuracy levels as in the case of fuzzy.