The paradigm of teaching ANN with an “interval teacher” is considered to solve the problems of medical diagnosis of chronic kidney disease. In this concept, the ANN implemented a mechanism for choosing a specific “teacher” from the deterministic interval set. A comparison of the learning results of the classical scheme and the proposed paradigm is given. The effectiveness of the proposed learning concept is shown. The accuracy of the classifier with the “interval teacher” is 84.6%.
CITATION STYLE
Mirkin, E., & Savchenko, E. (2022). Use of the Method of Setting the Interval Target in the Problem of Synthesis of a Neural Network Classifier for Diagnosing Chronic Kidney Disease in Patients. In Springer Proceedings in Physics (Vol. 268, pp. 57–66). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-81119-8_6
Mendeley helps you to discover research relevant for your work.