No abstract

IntroductionModern criminal groups can commit malicious activities against the state and society not only in the real world, but also in a virtual environment (Internet, e-government). These activities usually have various purposes, including propaganda against the state and dissemination of information, which promote terrorism and shake the foundations of national and spiritual values. [1][2][3][4][5][6][7].The timely detection of this information in e-government is very important in ensuring the security of the state and society, which is one of the most pressing scientific-theoretical and practical problems of our time [6,7]. It is no coincidence that the problem of e-government security is specified as one of the most actual 13 research directions to be studied in the eGovRTD2020 project adopted by the European Commission [8].One of the main functions of e-government is to protect the citizens from possible impairment and violence. Linders [9] examined the evolution of citizen-state relations and concluded that the Internet, in particular e-government, is the most effective and affordable tool to inform about possible crimes, as well as to improve the relations between community members and law enforcement authorities. However, studies show that criminal groups "benefit" from this favorable environment as well, and this has become a source of great danger for the state and society. 9/11 is the best example. Further analysis has shown that the organized criminal group behind 9/11 planned and coordinated the entire act via the Internet. One can naturally conclude the virtual world is a very favorable environment for criminal groups to commit their evil deeds.Understandably, one of the important tasks of the modern state is to identify and analyze secret criminal networks operating via the Internet and e-state. This virtual space has a wide range of opportunities to communicate and coordinate operatively. Criminal group members can communicate via websites, e-mail, blogs, online chats, etc. In most cases, transmitted information consists of text. Therefore, analysis of the text transmitted through the virtual environment is essential to preventing possible terrorist acts and ensuring the security of e-government [10].Currently, text mining, the intelligent analysis of text gathered from various sources, is considered one of the most advanced and effective technologies in knowledge control [11].Another reason for the popularity of text mining and its wide range of applications is the dominance of text in the information, regardless of the information's sources. According to International Data Corporation analysts, about 80% of produced information contains text [12].

Nowadays, the improvement of governance, ensurance the security and the timely detection of propaganda against the government are major problems of e-government. Extraction of hidden social networks is one of the most actual problems in the term of government security. The extraction of hidden social networks operating against the state in e-government is one of the key factors to ensure the security in e-government. In the paper, a method has been proposed for extracting hidden social networks to improve management in e-government, prevent promotion against the government and ensure the security. In this approach, hidden social networks are extracted through the analysis of user's comments via opinion and text mining technologies.

Nowadays, improvement of governance, ensuring security and timely detection of propaganda against the government are major problems of e-government. The extraction of hidden social networks operating against the state in e-government is one of the key factors to ensure the security in e-government. In this article, a method has been proposed for extracting hidden social networks to improve e-government management, prevent promotion against the government and ensure the security. In this approach, hidden social networks are extracted through the analysis of user's comments via opinion and text mining technologies. The authors assume that all comments are written in one language. Unlike previous methods, to detect social relationships between actors, content analysis technology, namely opinion mining technology was used in the proposed approach.

E-government expresses the process of utilizing advanced information and communication technologies (ICT) to automate internal activities of government agencies and their external relations with citizens and businesses. All these interactions provide better, faster and more secure public services. In this article, a method for the detection of terrorism-related activities in the e-government environment has been suggested. In the proposed method, terrorism-related activities are defined based on the similarity between the users' opinions and the vocabulary database created linked to terrorism.