Institutional Repository of Vilnius University
Ryšio su vartotojais hiperpersonalizacija, paremta didelių duomenų analize ir dirbtinio intelekto modelių integravimu /
Abstract
dc:descriptionBrief description of the thesis: This thesis examines how hyper-personalization affects user experience through big data analysis and artificial intelligence models. The theoretical section reviews technological foundations and ethical challenges, while the methodological section employs quantitative research to analyse user experience and attitudes toward personalized online content. The research findings revealed a lack of awareness, trust deficit, and transparency issues. In the practical section, based on theoretical analysis and research results, a hyper-personalization transparency dashboard was designed and implemented, integrated into a “WooCommerce” e-commerce store, helping users understand and manage the data collection and hyper-personalization process. Research problem - Traditional personalization no longer meets individual user needs, while hyper-personalization, despite offering more advanced solutions, faces privacy, ethical, and technological challenges. This raises the question of how to effectively integrate big data analysis and AI models to improve user experience related to hyper-personalization. Research aim - To develop an information system aimed at improving user experience in the hyper-personalization context, based on big data analysis and artificial intelligence model integration. Objectives: 1. To review and systematize aspects of hyper-personalization, big data, and artificial intelligence models examined in scientific literature, and to investigate ethical and privacy challenges posed by hyper-personalization according to transparency, awareness, and trust criteria. 2. To conduct empirical research to determine users' levels of awareness, trust, and discomfort, and to summarize key needs in the hyper-personalization context. 3. To substantiate the proposed hyper-personalization transparency dashboard solution, develop a detailed system specification, and formulate functional and non-functional requirements. 4. To design the architecture of the hyper-personalization transparency dashboard and its main system components. 5. To implement and test the designed hyper-personalization transparency dashboard in an e-commerce environment, evaluating its functionality and compliance with established requirements. Research methods applied: Literature analysis, systematization and comparison, empirical research. Research conducted and results obtained: Quantitative research results showed that most respondents frequently encounter hyper-personalized content, especially on social networks and in online stores. However, as many as 94.2% of respondents experienced discomfort due to such content, primarily due to lack of transparency and uncertainty about collected data. Respondents indicated that they lack information and control options, which raises distrust in hyper-personalization solutions. Conclusions: The theoretical analysis and empirical research results demonstrated that while hyper-personalization enables more precise and effective content adaptation to users, its success requires addressing issues related to privacy, transparency, and awareness. In response to the identified challenges, a hyper-personalization transparency dashboard was designed and implemented, integrated into a WooCommerce e-commerce store, which provides real-time information to users about data collection, visualizes the hyper-personalization process, and enables management of hyper-personalization settings. Keywords: hyper-personalization, big data, artificial intelligence, user experience, data privacy, data transparency.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Litvaitytė, Ignė,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD277692229&prefLang=en_US