Voume 3 : Issue 1

A Pragmatic Review of Data Mining Techniques for Thyroid Categorization

Authors : Poornima D.,Asha Gowda Karegowda.



Medical data mining has great potential for exploring the hidden patterns in broad variety of data sets, including medical domain. The foremost focal point of our analysis is to study and compare the performance of data mining techniques namely: supervised learning, unsupervised learning and association rule mining for the diagnosis of Thyroid disorders. Furthermore, applications of ensemble learning and fuzzy inference system for Thyroid disorders detection are also investigated. In addition, this paper brings out a range of feature selection methods applied for enhancement of classification of Thyroid disorders. The study reveals that, supervised classification has been extensively used when compared to unsupervised and association rule mining. Among the various supervised algorithms, it was observed that neural network has been expansively used for the Thyroid detection problem.