Hetero Associative Memory Based Neural Network Classifier for Health Care Data Diagnosis
R. Akila1, M. Nandhini2, S N Sivanandam3
1R. Akila, Computer Science and Engineering, PSG College of Technology Coimbatore.
2Asst. Prof M. Nandhini, Computer Science and Engineering, PSG College of Technology, Coimbatore.
3Dr. S N Sivanandam, Computer Science and Engineering, Karpagam College of Engineering, Coimbatore.
Manuscript received on May 05, 2015. | Revised Manuscript received on May 09, 2015. | Manuscript published on May 15, 2015. | PP: 86-89 | Volume-3 Issue-6, May 2015. | Retrieval Number: F0875053615/2015©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Classification is one of the predictive data mining tasks used to discover a model from the past data to predict some response of interest. In this work, Hetero Associative Memory based Neural Network (HAMNN) classifier is employed for health care data diagnosis. Classifier performance is enhanced by using Lern matrix, a popular model for associative memory. HAMNN classifier is built efficiently to improve the classification accuracy. This classifier provides promising results when experiments were conducted using six health care datasets from UCI machine learning repository.
Keywords: Neural network, Associative memory, Hetero associative memory, Lern matrix.