Scalable visualization for DBpedia ontology analysis using Hadoop

Abstract

Summary As ontologies are becoming larger and more diverse, ontological analysis and visualization of results have become more challenging, rendering the need for more computing resources. To address this issue, we suggest a system based on Hadoop for ontological analysis for large ontologies. Our suggested system consists of three parts: a data server to analyze ontological data, a visualization server to visualize the result of data analysis, and user applications to provide users with the visualized data. Server applications are implemented based on the Hadoop framework, and the ontological data are processed efficiently using the MapReduce algorithm. We performed ontological analysis using the DBpedia ontology and visualized the result. The goal of the visualization process is to determine the major properties of each class, and visualization is conducted on the Web in order to provide users with a cross‐platform environment. We evaluate the performance of the method by measuring execution times and analyzing experimental results obtained in the visualization process. The system we present is scalable for big ontological data. Copyright © 2015 John Wiley & Sons, Ltd.

Ontology specific visual canvas generation to facilitate sense-making-an algorithmic approach

“…For the verification of the suggested conceptualizations, visualization is a must. The screen size, on the other hand, serves as a permanent barrier, limiting understanding of visualized contents for both ontologists and domain experts [2]. Ontology development is a collaborative effort including ontologists and domain experts.…”

Visualisation for ontology sense-making: A tree-map based algorithmic approach

“…In ontological taxonomy or structure assessment 'visualization compactness 'is an insolvable problem, since the screen size will become a fixed barrier [1], [2]. However, visualization clarity can be enhanced via rational blend of appropriate visualization techniques [3].…”

Semantic framework for abstraction of things and applications in the future internet

“…The [112] investigates the management of large-scale RDF Graphs, proposing a Hadoop-based framework that will support functions for the store and the retrieval of the RDF data. In [113] it is presented the Hbase application [114] for the data storage and the MapReduce techniques for the deployment of a large-scale semantic data storage system. This technique is based on the Hexastore [115] application, combined with the RDF data modeling complemented with HBase presented in [116].…”