An Approach to Creating an Intelligent System for Detecting and Countering Inappropriate Information on the Internet

摘要

Currently, the Internet is becoming one of the most dangerous threats to personal, public and state information security. Therefore, the task of detecting and counteracting inappropriate information in digital network content becomes of national importance. The paper offers a new approach to creating an intelligent system for detecting and counteracting inappropriate information on the Internet based on the use of machine learning methods and processing of big data and describes the architecture of such a system. Experimental evaluation of one of the most important system components, which is the component of multidimensional evaluation and categorization of information objects in single-threaded and multi-threaded modes showed high efficiency of using various classifiers included in the Python Scikit-learn and Spark MLlib libraries to solve the problem.