Ontological Approach to Analysis of Big Data Metadata

Abstract

Now data becomes the most valuable commod- ity that allows to make informed decisions in various areas of human life. In this article, we look at the features of Big Data generated by the Internet of Things (IoT) technology, and also present the methodology for Big Data processing with use of semantic modeling (ontologies) at all stages of the Big Data life cycle. Use of Big Data semantic model allows eliminating such contradictions in these technologies as the heterogeneity of devices and things that causes the heterogeneity of the data types produced by them. Machine learning is used as an instrument for analyzes of Big Data: it provides logical inference of the rules that can be applied to processing of information generated by smart home system. In this methodology, the authors propose the use deep machine learning, based on convolutional neural networks because this model of machine learning corresponds to processing of unstructured and complex nature of the IoT domain. This approach increases the efficiency of IoT Big Data processing and differs from traditional processing systems by using NoSQL database, distributed architectures and semantic modeling. We propose the conceptual architecture of the Big Data processing system for IoT and describe it on example of the NoSQL database for the smart home. This architecture consists of five independent levels. A combined approach of semantic modeling and data mining methods can be used at each of these levels. Currently, this platform can be combined with a lot of open source components.