Title :
Conversational content in the context of safety of social networks
Author :
Stefa, Julinda ; Michalik, Peter
Author_Institution :
Fac. of Electr. Eng. & Inf. of Kossice, Tech. Univ., Kosice, Slovakia
Abstract :
The anticipated article concentrates on the conversational content of communication in social network. It defines the conversational content, its elaboration and analysis. It describes the present status of text-mining in communication content. It deals with the use of communication content for the identification of an author. As part of this contribution the issue of suitable applications and mining techniques of communication content is being elaborated as well. It describes in more detail the principle of Naïve Bayes Classification functioning within the given problem issue. It puts the chosen applications into coherence with theme modeling. In this context the interconnection of term and style based approach is being emphasized which by their mutual connection create an opportunity of theme identification and author identification of the conversational content for the sake of solution of the given problem.
Keywords :
Bayes methods; data mining; pattern classification; social networking (online); text analysis; author identification; communication content; conversational content; mining techniques; naïve Bayes classification; social network safety context; style based approach; term based approach; text-mining; theme identification; Computational modeling; Context; Dictionaries; Hidden Markov models; Safety; Social network services; Standards;
Conference_Titel :
Computational Cybernetics (ICCC), 2013 IEEE 9th International Conference on
Conference_Location :
Tihany
Print_ISBN :
978-1-4799-0060-2
DOI :
10.1109/ICCCyb.2013.6617576