Open AccessSYSTEM FOR CORRELATION ANALYSIS OF STATISTICAL INFORMATION ON CORONAVIRUS INCIDENCE
Author(s) -
I.M. Yannikov,
AUTHOR_ID,
M.V. Telegina,
A.N. Isenbaev,
E.N. Isenbaeva,
AUTHOR_ID,
AUTHOR_ID,
AUTHOR_ID
Publication year - 2021
Publication title -
izvestiâ samarskogo naučnogo centra rossijskoj akademii nauk
Language(s) - English
Resource type - Journals
ISSN - 1990-5378
DOI - 10.37313/1990-5378-2021-23-4-133-144
Subject(s) - computer science , relevance (law) , the internet , parsing , coronavirus , statistical hypothesis testing , correlation , data mining , data science , information retrieval , statistics , covid-19 , artificial intelligence , mathematics , world wide web , medicine , geometry , disease , pathology , political science , infectious disease (medical specialty) , law
The article is devoted to conducting a study on the influence of various sociological, economic, environmental and other factors on the state of the incidence and spread of coronavirus in the world. The authors proposed a scheme for obtaining information from Internet resources with the possibility of conducting a correlation analysis of data on the causes, rates and scale of the pandemic, and the factors affecting its spread. The introduction shows the relevance of the topic, carried out a detailed analysis of Internet resources. The systematization of the data placed in them has been carried out, the necessary conclusions and conclusions have been drawn. The website coronavirus-monitor.ru was selected as a source of statistical information. As a toolkit, it is proposed to develop the Coronavirus Stat program, which is necessary to view statistical data on a PC offline and carry out calculations to test hypotheses about the influence of external factors on the spread and course of COVID-19. The authors considered in detail the methods of obtaining information from Internet sources, their advantages and disadvantages, the method of automatic search with the development of their own version of the parser was chosen. Requirements for the selection of factors for correlation analysis and testable hypotheses are formulated. Examples of testing hypotheses with the presentation of graphs of dependences of the number of cases on various factors and correlation fields are given. A detailed description of the developed program, consisting of the frontend part of the program, a parser for obtaining new information, a database for storing old information, files for storing static information, is carried out. Requirements for the parser are formulated, a block diagram of its algorithm is presented. The requirements taken into account in the development of the program are shown and examples of its work are given. Testing of the program was carried out by conducting experiments to test the hypotheses put forward. The results of the experiments are summarized in the table. In conclusion, conclusions are drawn on the further use of the developed program.