{"title":"俄罗斯的战略管理:基于媒体数据自动构建因果关系图和识别关键主题","authors":"A. Zagranovskaia","doi":"10.37791/2687-0649-2023-18-2-73-84","DOIUrl":null,"url":null,"abstract":"Modern economic conditions are characterized by a high degree of uncertainty and complexity, which is difficult to formalize. Fuzzy cognitive maps make it possible to solve this problem – to cope with complexity, but when cognitive maps are built on the basis of expert opinions, this sometimes causes distrust due to the subjectivity of the judgments of individual specialists and doubts about compliance with the examination procedure. Therefore, the task of developing analytical tools to increase the awareness of decision makers about the real state of affairs in the organization and in the external environment is relevant, because contributes to the growth of their efficiency. The article proposes and tests a procedure for automated construction of a cause-and-effect diagram using statistical methods, as well as methods and models of machine learning. With the help of modern methods of topic modeling, key topics (concepts) are identified in the area under consideration for the considered period of time. The Doc2Vec model is then used to derive a fixed length numeric vector from the identified topics. The Granger test is then used to establish the possibility of a causal relationship between the topics found. The constructed cause-and-effect diagram allows you to describe the current situation and understand the key concepts of the area under consideration. According to the Russian media for 20 years (from 2002 to 2021), a cause-and-effect diagram was built that reflects the problems of strategic management in Russia. The analysis of the diagram made it possible to conclude that the topic of Russian projects is the most significant in the area under consideration.","PeriodicalId":44195,"journal":{"name":"Journal of Applied Mathematics & Informatics","volume":null,"pages":null},"PeriodicalIF":0.4000,"publicationDate":"2023-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Strategic management in Russia: automated construction of a cause-and-effect diagram and identification of key themes based on media data\",\"authors\":\"A. Zagranovskaia\",\"doi\":\"10.37791/2687-0649-2023-18-2-73-84\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Modern economic conditions are characterized by a high degree of uncertainty and complexity, which is difficult to formalize. Fuzzy cognitive maps make it possible to solve this problem – to cope with complexity, but when cognitive maps are built on the basis of expert opinions, this sometimes causes distrust due to the subjectivity of the judgments of individual specialists and doubts about compliance with the examination procedure. Therefore, the task of developing analytical tools to increase the awareness of decision makers about the real state of affairs in the organization and in the external environment is relevant, because contributes to the growth of their efficiency. The article proposes and tests a procedure for automated construction of a cause-and-effect diagram using statistical methods, as well as methods and models of machine learning. With the help of modern methods of topic modeling, key topics (concepts) are identified in the area under consideration for the considered period of time. The Doc2Vec model is then used to derive a fixed length numeric vector from the identified topics. The Granger test is then used to establish the possibility of a causal relationship between the topics found. The constructed cause-and-effect diagram allows you to describe the current situation and understand the key concepts of the area under consideration. According to the Russian media for 20 years (from 2002 to 2021), a cause-and-effect diagram was built that reflects the problems of strategic management in Russia. The analysis of the diagram made it possible to conclude that the topic of Russian projects is the most significant in the area under consideration.\",\"PeriodicalId\":44195,\"journal\":{\"name\":\"Journal of Applied Mathematics & Informatics\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.4000,\"publicationDate\":\"2023-03-31\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Applied Mathematics & Informatics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.37791/2687-0649-2023-18-2-73-84\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"MATHEMATICS, APPLIED\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Applied Mathematics & Informatics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.37791/2687-0649-2023-18-2-73-84","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"MATHEMATICS, APPLIED","Score":null,"Total":0}
Strategic management in Russia: automated construction of a cause-and-effect diagram and identification of key themes based on media data
Modern economic conditions are characterized by a high degree of uncertainty and complexity, which is difficult to formalize. Fuzzy cognitive maps make it possible to solve this problem – to cope with complexity, but when cognitive maps are built on the basis of expert opinions, this sometimes causes distrust due to the subjectivity of the judgments of individual specialists and doubts about compliance with the examination procedure. Therefore, the task of developing analytical tools to increase the awareness of decision makers about the real state of affairs in the organization and in the external environment is relevant, because contributes to the growth of their efficiency. The article proposes and tests a procedure for automated construction of a cause-and-effect diagram using statistical methods, as well as methods and models of machine learning. With the help of modern methods of topic modeling, key topics (concepts) are identified in the area under consideration for the considered period of time. The Doc2Vec model is then used to derive a fixed length numeric vector from the identified topics. The Granger test is then used to establish the possibility of a causal relationship between the topics found. The constructed cause-and-effect diagram allows you to describe the current situation and understand the key concepts of the area under consideration. According to the Russian media for 20 years (from 2002 to 2021), a cause-and-effect diagram was built that reflects the problems of strategic management in Russia. The analysis of the diagram made it possible to conclude that the topic of Russian projects is the most significant in the area under consideration.