No Detail Found!

Automated detection of destructive contents on the internet using data mining and machine learning methods

Author(s)
RYSKUL MEDETBEKOVA M.Auezov South Kazakhstan University, Shymkent, Kazakhstan
ZLIKHA MAKHANOVA M.Auezov South Kazakhstan University, Shymkent, Kazakhstan
BAYAN MYRZAKHMETOVA South Kazakhstan State Pedagogical University, Shymkent, Kazakhstan
AIGERIM TOKTAROVA South Kazakhstan State Pedagogical University, Shymkent, Kazakhstan; Khoja Akhmet Yassawi International Kazakh Turkish University, Turkistan, Kazakhstan
AIGUL IMANBAYEVA M.Auezov South Kazakhstan University, Shymkent, Kazakhstan
KADRZHAN SHIYAPOV Abai Kazakh National Pedagogical University, Almaty, Kazakhstan
GULNARA SEIDALIYEVA Kazakh National Agrarian Research University, Almaty, Kazakhstan
GAUKHAR SEIDALIYEVA University of International Business, Almaty, Kazakhstan
NURBEK KONYRBAEV Korkyt Ata Kyzylorda Ubiversity, Kyzylorda, Kazakhstan
RUSTAM ABDRAKHMANOV International University of Tourism and Hospitality, Turkistan, Kazakhstan
Abstract
In the context of ensuring the information and psychological security of society, the problem of uncontrolled growth of destructive content on the Internet is considered. Solving the problem of searching and identifying destructive information requires a minimum degree of subjectivity and maximum automation, since currently it is solved mainly by expert methods with the compilation of registers of prohibited sources. The existing search methods (expert processing, thematic search, intelligent data processing methods) used to find specific words in the text are considered, the advantages and disadvantages of these methods are noted. To solve this problem, methods of data mining and machine learning are proposed to search for destructive content using the example of profanity in textual information. The proposed approach is distinguished by the possibility of self-completion of the dictionary by a system based on the judgment of the identification of an unfamiliar word classified as non-normative. The correctness of the dictionary replenishment is measured using different metrics. A detailed description of this study is presented, starting from data collection to classification and detection of destructive content. An example of the algorithm and the initial results of its testing are given.
Publication Details
Page(s) 2358-2371
DOI DOI not available
Published Journal: Journal of Theoretical and Applied Information Technology, Volume: 100, Issue: 7, Year: 2022

Request the full text from PASTIC

Keywords
machine learning, Security, Internet, detection, Natural Language Processing., Destructive Data
0

Citations

0

Downloads

52

Views