Framework of integrated big data: A review

Abstract

Currently, how to deeply distill potential attributes of big data has become a great challenge for structured, semi-structured and unstructured data (SSU data) with a unified model. Structured data refers to any data that resides in a fixed field within a record or file including data contained in relational databases and spreadsheets. Unstructured data refers to data from text, pictures, audio, video, and other sources that do not fit into a relational database. Semi-structured data is information that doesn't reside in a relational database but that does have some organizational properties that make it easier to analyze, such as XML, and HTML documents. In this paper, we present a literature survey and a framework, namely integrated big data (IBD), which aims at exploring the approaches for constructing a universal IBD model, including representation, storage and management, computation, and visual analysis. Firstly, we present a systematic framework to decompose big data analytics into four modules. Next, we present a detailed survey of numerous approaches for these four modules. The main contributions of this paper are summarized in two dimensions. First, we propose a novel integrated big data framework for unified big data representation, storage, computation, and visual analysis. Second, we present the possible future methods in realizing the framework by reviewing methods. Through this paper, we would like to point out a promising research direction in unified investigation and application of big data.