Text mining to Discover Design Features for Cybersecurity tools: The case of Password Management Systems

Abstract

Username and password are the most common authentication approach used today in cybersecurity. Most people have more than one online account protected by passwords. Managing passwords is one of the daunting tasks for most people. Therefore, users may rely on password management systems PMS to help creating, managing, retrieving, and storing passwords. However, there is still a lack of relying on and using such systems to manage users’ passwords. This study aims to use a text mining approach to investigate the factors that influence users’ behavior to use PMSs which leads to improve their security. This study adopts a text mining approach to uncover the design principles (features) of PMSs extracted from online users’ reviews and feedbacks to improve the adoption of PMSs. Specifically, our study used a topic modeling algorithm to analyze users’ feedback by collecting and analyzing the reviews of nine password management systems. The results indicate the need to address principles and design features related to compatibility, security, usefulness, ease of use, and customer care services.