The Application of Semantic-based Classification on Big Data

Sensory networks are scale-free environments that connect entities remotely but with noticeable tendency among its participating Sensors. The increasing size of data sets and the lack of algorithmic methods that are effectively manage such huge data collections led to growing demands of new techniques to handle big data's side-effects. In this research, a new ontology-based categorization methodology is proposed. The novelty of this research stems for its focus on modularizing the classification task into multi-layer framework to group data in sensory networks. The three-layer framework assumes that large datasets of sensory networks are heterogeneous. Therefore, an ontology-layer could be created to identify semantic interpretation of data and semantic relationships with other domains' data. The goal of this research is to provide a technique that facilitates extracting ontological patterns, which enhance the semantic interpretation of such pool of knowledge. Furthermore, the proposed framework facilitates integrating different heterogeneous sources of knowledge into a single one.