This article is cited in 1 paper

System analysis, management and information processing

Construction of Kohonen self-organizing map (SOM) for prediction of mudflow types

R. A. Zhilov

Institute of Applied Mathematics and Automation – branch of Kabardino-Balkarian Scientific Center of the Russian Academy of Sciences, 360000, Russia, Nalchik, 89 A Shortanov street

Abstract: The paper describes a self-organizing Kohonen map (SOM) that analyzes the mudflow type. SOM is trained on real cadastre data of mudflow danger in the south of the European part of Russia. The purpose of the work is to obtain forecasts of mudflow types. The results of the work show that SOM provides good accuracy in predicting mudflow types. The main task will be to cluster data related to geological and meteorological factors in order to identify patterns that can be used to predict the risk of occurrence of various mudflow types. It is expected that the results of this work will contribute to more accurate and on time forecasting of mudflows, which, in turn, will help minimize damage from these natural phenomena.

Keywords: data clustering, SOM clustering method, SOM model, Kohonen self-organizing maps, mudflow type classification

UDC: 519.7

Received: 12.08.2024
Revised: 16.09.2024
Accepted: 23.09.2024

DOI: 10.35330/1991-6639-2024-26-5-129-137

Bibliographic databases:

•