Potenziale Künstlicher Intelligenz zur Optimierung der User Experience im E-Commerce

Abstract

(i) Motivation and problem definition The rapid development of artificial intelligence and the increasing spread of the technology leads to the question of the extent to which AI technologies can help to optimise the user experience in e-commerce. Traditional methods of personalisation are not capable of capturing complex user needs and situational contexts. However, with increasing competition in retail, the relevance of outstanding user experience is growing and is becoming an important differentiating factor. The thesis therefore deals with the question of how AI can contribute to optimising the user experience. (ii) Content structure and methodology A systematic literature review was conducted to answer the central topic of the thesis. Useful literature was defined with the help of inclusion and exclusion criteria. The theoretical foundations of UX, UI, AI and e-commerce were analysed with the help of further fundamental literature. Relevant machine learning and deep learning technologies were then analysed in terms of their functional significance and their potential for use in UX design. Based on this, application-oriented possibilities were presented, including AI-supported behavioural analyses and adaptive UI systems. (iii) Concrete results of the work The work showed that the AI technologies, machine learning and deep learning, as well as their sub-technologies, can be used effectively to identify and interpret user behaviour. With the help of reinforcement learning, recommendation systems can be dynamically adapted to real-time behaviour. So-called transformer models enable the context-bound analysis of sequential data. Generative adversarial networks help to optimise the design of visual elements and the completion of incomplete data sets. The work also shows that emotional and psychological states of users can be recognised using deep learning and used for UX optimisation. It is obvious that artificial intelligence has great potential for improving the user experience in e-commerce, but there are also challenges. There is a need for optimisation, particularly regarding the lack of transparency in the algorithmic decisions of complex models. Overall, however, the work shows that AI technologies have great potential for the usercentred optimization of the user experience in e-commerce.

Date of Award	2025
Original language	German (Austria)
Supervisor	Armin Johann Schnürer (Supervisor)