Decision trees have been widely used for classification in Data mining. Number of decision tree algorithms has been developed in the past. In order to reduce the computational time the On Improving the efficiency of SLIQ (OIESLIQ) algorithm has been developed with an aim to reduce diversity of the decision tree at each split. In order to improve the accuracy, the paper proposes a novel approach (Pioneer classifier algorithm) to embark upon the other two algorithms.

Discovering the information about several disorders prevailing in brain and neurology is by no means a new scientific technique. A neurological disorder of any human being can be analyzed using EEG (Electroencephalography) signal from the electrode's output. Epilepsy (spontaneous recurrent seizure) detection is usually carried out by the physicians using a visual scanning of the signals produced by EEG, which is onerous and may be inaccurate. EEG signal is often used to determine epilepsy, for its merits, such as non-invasive, portable, and economical, can exhibit superior temporal tenacity. This paper surveys the existing artifact removal methods. It puts a new-fangled mode forward to confiscate artifacts and hauls informative derived values from EEG to automate Epilepsy detection. The automated Epilepsy detection has to precisely indicate and detect the neural abnormality of the brain. This indication and detection process necessitates a proficient approach for the prompt removal of artifacts of the EEG signals. An effective artifact removal of EEG signals can alone enable the useful features of the original signals for further processing. Once the original signals excluding the noise is obtained, a delicate strategy for extracting the features of the signals, becomes mandatory in order to accomplish robust classification of the signal. Then an expert classification technique is implemented to aid the automated analysis process to correctly distinguish the EEG signal features.