ALGORITHMS FOR PREDICTING INFORMATION SECURITY VULNERABILITIES AND THREATS BASED ON THEMATIC INTERNET RESOURCES

Abstract

The considered features of the functioning of forums of thematic Internet resources allow to forecast the occurrence of vulnerabilities and threats to the security of confidential data. To solve this problem, the analysis of Internet messages created by participants of forums of thematic Internet resources is expected, according to the presented algorithms.

The algorithm for predicting vulnerabilities and threats to information security is distinguished by the possibility of detecting vulnerabilities and threats at the early stages of their practical implementation, is based on the analysis of the flow of messages of forums of thematic Internet resources, which allows information security specialists to take adequate and timely measures to protect confidential data. The result of the algorithm's work is reports on identified vulnerabilities and threats to the information security of confidential data. The reports may also include information reflecting the results of text message analysis, based on which a conclusion about the occurrence of vulnerabilities and threats was obtained.

The message flow filtering and statistical analysis algorithm involves filtering thematic messages that do not belong to the given subject area, which is specified by the corresponding ontology. The result of the algorithm is the determination of statistical indicators characterizing the flow of thematic messages during the period of data flow analysis. The algorithm also calculates the number of text messages that have passed the data flow filtering stage and determines the average rating of text message authors. The results of the algorithm can be used to build a system of logical fuzzy inference for forecasting the events of the subject area for which the analysis is conducted.

The proposed algorithms allow predicting vulnerabilities and threats, taking adequate measures to protect information. The obtained results testify to the effectiveness of the proposed algorithms for forecasting vulnerabilities and threats, and also confirm the correctness of the information and analytical system.