Empirical evaluation of Linked Data visualization tools

• Published in: Future generation computer systems Vol. 112; pp. 258 - 282
• Main Authors: Desimoni, Federico, Po, Laura
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.11.2020
• Subjects: Linked Data; Linked data exploration guidelines; Open data; Visualization tools; Linked Data; Open data; Visualization tools; Linked data exploration guidelines
• ISSN: 0167-739X; 1872-7115
• DOI: 10.1016/j.future.2020.05.038

The economic impact of open data in Europe has an estimated value of € 140 billions a year between direct and indirect effects. The social impact is also known to be high, as the use of more transparent open data have been enhancing public services and creating new opportunities for citizens and organizations. We are assisting at a staggering growth in the production and consumption of Linked Data (LD). Exploring, visualizing and analyzing LD is a core task for a variety of users in numerous scenarios. This paper deeply analyzes the state of the art of tools for LD visualization. Linked Data visualization aims to provide graphical representations of datasets or of some information of interest selected by a user, with the aim to facilitate their analysis. A complete list of 77 LD visualization tools has been created starting from tools listed in previous surveys or research papers and integrating newer tools recently published online. The visualization tools have been described and compared based on their usability, and their features. A set of goals that LD tools should implement in order to provide clear and convincing visualizations has been defined and 14 tools have been tested on a big LD dataset. The results of this comparison and test led us to define some suggestions for LD consumers in order for them to be able to select the most appropriate tools based on the type of analysis they wish to perform. •An in-depth analysis of the main linked data visualization tools available is reported.•77 linked data visualization tools have been discovered and tested.•29 working tools have been compared based on their features regarding visualization types, entry point, query capabilities, type of visualizations and export types.•10 goals that LD visualization tools should implement in order to provide clear and convincing visualization of the data contained in an LD source has been formalized.•An evaluation of the behavior of LD tools on a big LD dataset w.r.t. the defined goals has been carried out and has let us to draw some conclusion on the different LD tool that are actually available.