WSEAS Transactions on Information Science and Applications
Print ISSN: 1790-0832, E-ISSN: 2224-3402
Volume 10, 2013
An Optimization of a Planning Information System Using Fuzzy Inference System and Adaptive Neuro-Fuzzy Inference System
Authors: ,
Abstract: This paper aims to design Mamdani-Fuzzy Inference System(FIS) and Sugeno-Adaptive Neuro-Fuzzy Inference System model (ANFIS) for the development of an effective Information System. The comparative study of both the systems provided that the results of ANFIS model are better than the Fuzzy Inference System [6]. This was ascertained after testing of the models. Sugeno-type ANFIS has an advantage that it is integrated with neural networks and genetic algorithm or other optimization techniques therefore the IS Planning Model using ANFIS adapt to individual user inputs and environment. In this research paper, the datasets loaded into FIS and ANFIS have the responses of 86 managers regarding the factors responsible for the success, challenge and failure of Information System. A Fuzzy Inference System has been designed having input fields of various factors under three sub dimensions (strategic planning, top management and IS infrastructure) responsible for the Information System Planning whereas output fields are success, challenge or failure of IS planning. In the development of ANFIS model for IS planning, input values of various planning factors like strategic planning, top management and IS infrastructure determine the success/challenged/failure of IS planning. Comparison of two systems (FIS and ANFIS) results shows that the results of ANFIS model are better than FIS when these systems designed and tested. In ANFIS model, for IS planning output fields, the training, testing and checking errors are 0.0204, 0.4732 and 0.27607 where as FIS results in average error of 4.87.
Search Articles
Keywords: Information System (IS), Critical Success Factors (CSFs), Critical Failure Factors (CFFs), ANFIS, FIS, Planning Information System