Semantic data discovery from Social Big Data

TL;DR

This paper explores semantic analytics and domain knowledge modeling to improve understanding and management of social big data, addressing challenges like data heterogeneity, streaming, and ambiguity.

Contribution

It provides an overview of semantic analysis, domain ontology, and knowledge graphs, demonstrating their application in social data analysis through a case study.

Findings

Semantic analytics enhances social data understanding.

Knowledge graphs interlinking improves data retrieval.

Case study validates the approach in political social data.

Abstract

Due to the large volume of data and information generated by a multitude of social data sources, it is a huge challenge to manage and extract useful knowledge, especially given the different forms of data, streaming data and uncertainty and ambiguity of data. Hence, there are still challenges in this area of BD analytics research to capture, store, process, visualise, query, and manipulate datasets to derive meaningful information that is specific to an application's domain. This chapter attempts to address this problem by studying Semantic Analytics and domain knowledge modelling, and to what extent these technologies can be utilised toward better understanding to the social textual contents. In particular, the chapter gives an overview of semantic analysis and domain ontology followed by shedding light on domain knowledge modelling, inference, semantic storage, and publicly available semantic tools and APIs. Also, the theoretical notion of Knowledge Graphs is reported and their interlinking with SBD is discussed. The utility of the semantic analytics is demonstrated and evaluated through a case study on social data in the context of politics domain.