TY  - JOUR
T1  - Enhanced Learning Approach for Diseases Diagnostic
AU - Fouad, Khaled M. AU - Shishtawy, Tarek El AU - Altae, Aymen A. 
JO  - Asian Journal of Information Technology
VL  - 17
IS  - 3
SP  - 202
EP  - 211
PY  - 2018
DA  - 2001/08/19
SN  - 1682-3915
DO  - ajit.2018.202.211
UR  - https://makhillpublications.co/view-article.php?doi=ajit.2018.202.211
KW  - Fast Learning Network (FLN)
KW  -Extreme earning Machine (ELM)
KW  -(mRMR) algorithm
KW  -hybrid method
KW  -medical diagnosis
KW  -diseases
AB  - Medical disease diagnostic is a very important problem in the medical domain and data mining approach. Early detection of the diseases is very highly important for treating it in early stages. The challenges among machine learning methods are very important to focus on the effective tool to improve the diagnoses problem by indicating the performance of neural network classifiers. This research aims to create a new hybrid method (mRMR-FLN) by exploiting the potential performance of Fast Learning Network (FLN) classifier after integrating it with efficient feature selection algorithm, maximum Relevance Minimum Redundancy (mRMR), to achieve better diagnoses on different diseases. The components of the proposed hybrid method (mRMR-FLN) will be (mRMR) algorithm as a feature selection method and Fast Learning Network (FLN) as a neural classifier. The performance of the new model has been examined and recorded with benchmark measurement on seven evaluation measures. The proposed hybrid method (mRMR-FLN) has achieved very promising classification accuracy using 10-fold Cross-Validation (CV).
ER  -