Big Data analytics ontology

Autor: Vasyl Lytvyn, Victoria Vysotska, Oksana Oryshchyn, Oksana Brodyak, Oleh Veres
Jazyk: angličtina
Rok vydání: 2017
Předmět:
Zdroj: Tehnologìčnij Audit ta Rezervi Virobnictva, Vol 1, Iss 2(39), Pp 16-27 (2017)
ISSN: 2312-8372
2226-3780
Popis: The object of this research is the Big Data (BD) analysis processes. One of the most problematic places is the lack of a clear classification of BD analysis methods, the presence of which will greatly facilitate the selection of an optimal and efficient algorithm for analyzing these data depending on their structure. In the course of the study, Data Mining methods, Technologies Tech Mining, MapReduce technology, data visualization, other technologies and analysis techniques were used. This allows to determine their main characteristics and features for constructing a formal analysis model for Big Data. The rules for analyzing Big Data in the form of an ontological knowledge base are developed with the aim of using it to process and analyze any data. A classifier for forming a set of Big Data analysis rules has been obtained. Each BD has a set of parameters and criteria that determine the methods and technologies of analysis. The very purpose of BD, its structure and content determine the techniques and technologies for further analysis. Thanks to the developed ontology of the knowledge base of BD analysis with Protege 3.4.7 and the set of RABD rules built in them, the process of selecting the methodologies and technologies for further analysis is shortened and the analysis of the selected BD is automated. This is due to the fact that the proposed approach to the analysis of Big Data has a number of features, in particular ontological knowledge base based on modern methods of artificial intelligence. Thanks to this, it is possible to obtain a complete set of Big Data analysis rules. This is possible only if the parameters and criteria of a specific Big Data are analyzed clearly.
Databáze: OpenAIRE
Externí odkaz: