Emil Häglund defends his doctoral thesis, Contextual Intelligence: Leveraging AI for Targeted Marketing, at Umeå University.
Click here for a link to the event page at Umeå University
Abstract
As privacy concerns increase and regulation against tracking-based advertisingtightens, contextual advertising—which targets ads based on webpage content ratherthan personal data—offers a compelling alternative. The shift towards this alternative form of ad targeting is gaining momentum thanks to advancements in artificial intelligence (AI), which significantly improve the ability to interpret and categorize online content. This thesis explores how AI can interpret online contexts and leverage them for targeted, privacy-conscious marketing. A key contribution is the development of methods for extracting opinions from text and structuring them into “opinion units”, leveraging the power and versatility of large language models. Opinion units consist of concise, context-rich excerpts that capture individual opinions, paired with sentiment metadata. The proposed methods demonstrate high accuracy in opinion extraction and show promise for downstream applications. For instance, in opinion search and topic modeling of customer reviews, the compactness and distinctness of opinion units enhance retrieval precision and produce more coherent and interpretable groupings of opinions. This enables theidentification of specific aspects driving customer satisfaction, providing insights forproduct development and targeted marketing. Marketing experiments conducted in this thesis reveal how media contexts influence advertising perceptions. The findings demonstrate that engaging content and the credibility of website sources create a spillover effect, enhancing the effectiveness of associated ads. Regarding brand safety—ensuring ads do not appear in brand-damaging contexts—the results suggest that proximity to negative news articles alone is not directly harmful. However, marketers face increased risks when the advertised message is associated with a negative context. To mitigate these risks, AI tools can be used to detect and avoid potentially unsafe online environments. Finally, the thesis offers guidance on AI-driven ad targeting by outlining the trade-offs between contextual and personalized strategies, as well as manual versus automated methods. The discussion considers key factors such as marketing objectives, data availability, and ethical considerations alongside regulatory requirements. The findings serve as a foundation for making well-informed, strategic choices in the future of advertising targeting.
Supervisor
Johanna Björklund, Associate Professor at Umeå University
Opponent
Bernard Jansen, Professor at Qatar Computing Research Institute.
