Detection and Evaluation of Dark Patterns in E-business Websites Utilizing Deep Learning

With the rapid progress of Internet technology, e-business, and Internet finance have ushered in unprecedented development opportunities. However, a negative design phenomenon, known as "Dark Patterns" quietly emerged during this process. In the booming digital economy and online marketing environment, Dark Patterns specifically refer to those behaviors that deliberately induce users to make choices that are not in their best interests or have not been carefully considered, using visual misleading, complex interaction processes, or covert design, to seek additional benefits for enterprises or online platforms. This behavior not only damages the legitimate rights and interests of users, but also disrupts the fair competition order in the market. In response to this challenge, this article proposes a Dark Patterns detection and evaluation model based on deep learning (DL) technology. This model can intelligently identify and evaluate the existence and impact of various Dark Patterns by deeply mining user behavior data, interface design features, and subtle clues in the interaction process. The results show that the model performs well in recognition accuracy, false positive rate, and evaluation consistency, effectively verifying its feasibility and reliability in practical applications.