Relevant features selection is become primary preprocessing step for building
almost intelligence machine learning systems. Feature Selection (FS) is more and
more important in many applications such as patterns recognition, medical
technologies, data mining environments and others. The main objective of FS is to
choice the important features among multi set in order to building effective machine
learning models such as pattern analysis model by cancelling irrelevant or redundant
attributes. An addition to that, there is a fact that the efficiency of the desired system
is very sensitive to choose of the features that effect on classification or any analysis
procedure of small or high dimensional datasets. Furthermore, the analysis of medical
datasets has become growing claiming problem, due to huge datasets that cause time
consuming and uses additional computational effort, which may not be suitable for
many applications.
This work attempts to introduce a hybrid genetic k-means of feature selection
algorithm for multi medical diseases datasets. The proposed algorithm uses a genetic
algorithm combine with k-means algorithm as a powerful tool to select the relevant
features from different large medical datasets of Mirjan hospital diabetes, heart and
breast cancer diseases which play the important role in maximum the classification
accuracy and efficiency of the system. Experimental results show the efficiency of the
proposed system for the used datasets and satisfy maximum classification accuracy
performance compared with others states.