Visualization System for Transparency Requirement Analytics

Abstract

Access to corporate information systems by consumers via the Internet has increased dramatically over the past several decades. In a separate organization, extensive research has been conducted on the free flow of information generated by both external and internal keywords. Research on transparency should aid the audience in making informed decisions. Few have, however, created clear and compelling visual representations of transparency requirements (stakeholders, data, process, policy, and their relationships) utilizing current information visualization and visual analytics methodologies. Maintaining both the quality and visual representation of transparency requirements is a difficult challenge. In this paper, we propose TranspVis, a new visual analytics tool designed for transparency analytics. It consists of multiple views that aid domain experts in efficiently analyzing, updating, and saving application transparency datasets. TranspVis is an interactive tool for displaying TranspLan (i.e., Transparency Language) representations manually generated by experts utilizing the Shield, Infolet, and SitReq forms. In addition to the new circle view, TranspVis generates and synchronizes these latter representations automatically. TranspVis is evaluated using AWS and WhatsApp policy datasets as two case studies. Results show that TranspVis extends the initial TranspLan representation and significantly improves transparency requirement analytics in terms of visual encoding, interactions, and insight extraction.

Keywords

• information visualization
• visual analytics
• transparency analytics
• TranspLan modeling language
• data science