Soft computing for intelligent data analysis

Abstract

Intelligent data analysis (IDA) is an interdisciplinary study concerned with the effective analysis of data. The paper briefly looks at some of the key issues in intelligent data analysis, discusses the opportunities for soft computing in this context, and presents several IDA case studies in which soft computing has played key roles. These studies are all concerned with complex real-world problem solving, including consistency checking between mass spectral data with proposed chemical structures, screening for glaucoma and other eye diseases, forecasting of visual field deterioration, and diagnosis in an oil refinery involving multivariate time series. Bayesian networks, evolutionary computation, neural networks, and machine learning in general are some of those soft computing techniques effectively used in these studies.