Showing 1–1 of 1 results for author: Tverdokhlib, T

Hybrid clustering-classification neural network in the medical diagnostics of reactive arthritis

Authors: Yevgeniy Bodyanskiy, Olena Vynokurova, Volodymyr Savvo, Tatiana Tverdokhlib, Pavlo Mulesa

Abstract: The hybrid clustering-classification neural network is proposed. This network allows increasing a quality of information processing under the condition of overlapping classes due to the rational choice of a learning rate parameter and introducing a special procedure of fuzzy reasoning in the clustering process, which occurs both with an external learning signal (supervised) and without the one (un… ▽ More The hybrid clustering-classification neural network is proposed. This network allows increasing a quality of information processing under the condition of overlapping classes due to the rational choice of a learning rate parameter and introducing a special procedure of fuzzy reasoning in the clustering process, which occurs both with an external learning signal (supervised) and without the one (unsupervised). As similarity measure neighborhood function or membership one, cosine structures are used, which allow to provide a high flexibility due to self-learning-learning process and to provide some new useful properties. Many realized experiments have confirmed the efficiency of proposed hybrid clustering-classification neural network; also, this network was used for solving diagnostics task of reactive arthritis. △ Less

Submitted 21 October, 2016; originally announced October 2016.

Journal ref: International Journal of Intelligent Systems and Applications, 2016, Vol. 8, No. 8, pp.1-9