Clustering of Dark Patterns in the User Interfaces of Websites and Online Trading Portals (E-Commerce)

Abstract

dark patterns in the interfaces of users using sites and portals of online trading affect their behavior by companies that own digital resources. The authors propose to implement the detection of dark patterns on sites in user interfaces using cluster analysis algorithms using two methods for clustering many dark patterns in application interfaces: hierarchical and k-means. The complexity of the implementation lies in the lack of datasets that formalize dark patterns in user interfaces. The authors conducted a study and identified signs of dark patterns based on the use of Nelsen’s antisymmetric principles. The article proposes a technique for assessing dark patterns using linguistic variables and their further interval numerical assessment for implementing cluster data analysis. The last part of the article contains an analysis of two clustering algorithms and an analysis of the methods and procedures for applying them to clustering data according to previously selected features in the RStudio environment. We also gave a characteristic for each resulting cluster.