Rough set theory and rule induction techniques for discovery of attribute dependencies in medical information systems

Abstract

A b s t r a c t . Problems connected with applications of the rough set theory to identify the most important attributes and with induction of decision rules from the medical data set are discussed in this paper. The medical data set concerns patients with multiple injuries. The direct use of the original rough set model leads to finding too many possibilities of reducing the input data. To solve this difficulty, a new approach integrating rough set theory, rule induction and statistical techniques is introduced. First, the Chi-square test is additionally performed in order to reject non-significant attributes. Then, starting from remaining attributes we try to construct such definitions of new attributes that improve finally discovered decision rules. The results have shown that the proposed approach integrating all methods has given better results than those obtained by applying the original rough set method. I n t r o d u c t i o nOne of the essential issues in the analysis of experience represented in data sets is looking for dependencies among attributes describing objects of our interest. It is usually assumed that objects are described by two kinds of attributes: condition (features characterizing object properties by means of numerical and symbolic values) and decision (referring to decision, diagnosis and, in general, expressing classification of objects) attributes.In this paper, we consider the discovery of attribute dependencies as two subtasks:discovery of redundant attributes in the representation of dependencies between condition and decision attributes and identification of the most important condition attributes, -discovery of decision rules characterizing the dependency between values of condition attributes and decision attributes.In recent years the authors tried to solve the above tasks in the analysis of various data sets, mainly medical ones, using several data mining approaches

Soft Computing Under Uncertain Knowledge

“…A prediction system according to which the risk of Glaucoma can be assessed was developed by Alcantud et al [19]. Çelik and Yamak [30] proposed fuzzy soft sets in medical diagnosis, Stefanowski and Slowinski [91] show applications of rough sets to identify the causal relevancy of particular pre-therapy attributes. Alcantud et al [19] carried out an analysis of survival for lung cancer resections cases with fuzzy and soft set theory in decision making.…”

Analysis of survival for lung cancer resections cases with fuzzy and soft set theory in surgical decision making

“…Stefanowski and Slowiński applied rough set theory in order to pinpoint the most relevant parameters which are connected with the induction of decision rules from medical databases. In [73], these authors calculated a strong positive causal effect of particular pre-therapy attributes by specifying the preciseness with which patients are classified according to their specific recuperation. The use of soft set theory in the diagnosis of risk of prostate cancer by Yuksel et al .…”

Randomized controlled trials versus rough set analysis: two competing approaches for evaluating clinical data

“…The possibility of specifying the accuracy with which patients are assigned to particular recovery classes also allows one to determine the causal relevancy of particular pre-therapy attributes [ 16 ]. The idea is as follows.…”