Abstract
Medical data is heterogeneous in nature and associated with uncertainties. For that reason, data mining has been assisting physicians in decision making and to cope with the information overload. A considerable amount of literature has been available on medical data classification based on data mining techniques to automate or facilitating the delineation of images. However, from image formation to the final analysis, medical imaging is still facing challenges. New imaging procedures for classification could overcome the inefficiencies and provide more reliable information to the medical experts. Therefore, this paper assesses the performance of selected classification algorithms based on fuzzy soft set for classification of medical data. There are two concepts that underlie the classification in the fuzzy soft set theory namely: classification based on decision making problem and classification based on similarity between two fuzzy soft set. The selected algorithms are evaluated based on two criteria: accuracy and computational time. Moreover, the conducted experiments demonstrated the effectiveness of fuzzy soft set for medical data categorization.