Open AccessSystematic assessment of food security by recurrent addition of fuzzy cognitive maps
Author(s) -
Aleksey F. Rogachev
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1801/1/012027
Subject(s) - fuzzy cognitive map , set (abstract data type) , fuzzy logic , computer science , cognition , process (computing) , fuzzy set , strengths and weaknesses , management science , software , artificial intelligence , data mining , machine learning , industrial engineering , fuzzy number , engineering , psychology , social psychology , neuroscience , programming language , operating system
The results of an analytical review of the problems of modeling the state and evolution of socio-economic systems solved using the mathematical apparatus of matrix analysis of cognitive maps are presented. The difference between classical and fuzzy cognitive maps (FCM) is shown. The problems, strengths and weaknesses of algorithms and tools for assessing the level of food security (FS) based on a fuzzy cognitive approach are considered. The basic groups of concepts and methods for constructing the FCM that provide a fuzzy integral assessment of the PB level are analyzed. A method of recurrent construction of the NCC by successive addition of a set of concepts is proposed. In the process of adding concepts, experts make step-by-step recurrent adjustments to the values of the mutual influence of concepts. Recommendations for justification are presented, using the example of assessing the level of FS, the choice of tools for building the NCC. The solution of the problem of assessing the level of regional FS is illustrated for a set of economic, environmental and organizational concepts. It is shown that for adjusting the para-meters generated by the FCM, including to take account of their changes over time require the creation of software tools for fuzzy cognitive modeling, taking into account specific tasks and retrospective information together with simulated system factors.