Self organizing neural networks in prediction.

Venkatesan, P.; Mullai, M.

International Journal of Data Warehousing and Mining; 2013; 3; 175-179.

Abstract: Artificial Neural Network (ANN) has emerged as a powerful tool in handling uncertainty which is very essential in decision making especially in medical diagnosis. This performs in two different modes, supervised and unsupervised. In this paper performance of Self-Organizing map (SOM), the unsupervised category in prediction is compared with that of supervised category. Breast cancer data obtained from UCI -Machine Learning Repository is used for the purpose. Nine inputs have been given to the network and the network is trained with five as well as six nodes in the hidden layer. SOM is also used to do the same task and proved that its performance is commendable, inspite of its unsupervised nature. Matlab and Viscovery SOMine are used for the purpose.

Keywords: Medical diagnosis; Artificial Neural network (ANN); Self Organizing Map (SOM )

Back to List of publications / Home