Le travail de cette thèse s’inscrit dans le cadre de la recherche d’information (RI) et plus précisément la recherche d’information personnalisée. En effet, avec la prolifération des données de différentes sources et malgré la diversité de méthodes et d’approches de la recherche d’information classique, cette dernière n’est plus considérée aujourd’hui comme un moyen efficace pour répondre aux exigences de l’utilisateur considéré comme l’acteur principal de tout système de recherche d’information (SRI). Dans ce travail de thèse, nous adressons deux principaux problèmes liés à la RI personnalisée : (1) la formalisation et la mise en œuvre d’un modèle utilisateur et (2) la formulation de la requête de recherche dans le but d’améliorer les résultats retournés à l'utilisateur en fonction de sa perception et de ses préférences. Pour atteindre ces objectifs, nous avons proposé une approche de recherche d’information guidée par les ontologies et basée sur l’utilisation sémantique des informations. En effet, notre contribution se décline en trois principaux points : (1) la modélisation et la construction de profil utilisateur suivant une approche ontologique modulaire. Ce modèle permet de capturer les informations relatives à l’utilisateur, de les modéliser suivant l’approche sémantique dans le but de les réutiliser pour des tâches de raisonnement et d’inférence ; (2) la reformulation sémantique de la requête de recherche en exploitant les concepts, les relations syntaxiques et non syntaxiques entre les concepts et les propriétés ; et finalement, (3) la recommandation des résultats qui consiste à proposer des résultats de recherche en se basant sur l’ensemble des communautés utilisateurs construites par l’approche de classification non supervisée « Fuzzy K-mode » améliorée. Ces communautés sont aussi modélisées sémantiquement dans l’ontologie modulaire de profil. Ensuite, afin de valider l’approche proposée, nous avons mis en œuvre un système pour la recherche des itinéraires dans le transport public. Enfin, cette thèse propose des perspectives de recherche sur la base des limites rencontrées. / This PhD thesis reports on a recent study in the field of information retrieval (IR), more specifically personalized IR. Traditional IR uses various methods and approaches. However, given the proliferation of data from different sources, traditional IR is no longer considered to be an effective means of meeting users’ requirements. (‘Users’ here refers to the main actor in an IR system.) In this thesis, we address two main problems related to personalized IR: (1) the development and implementation of a user model; and (2) the formulation of a search query to improve the results returned to users according to their perceptions and preferences. To achieve these goals, we propose a semantic information search approach, based on the use of semantic information and guided by ontologies. The contribution of our work is threefold. First, it models and constructs user profiles following a modular ontological approach; this model allows the capture of information related to the user, and models the data according to the semantic approach so that the data can be re-used for reasoning and inference tasks. Second, it provides evidence for reformulating a query by exploiting concepts, hierarchical and non-hierarchical relationships between concepts and properties. Third, based on our findings, we recommend search results that are informed by the user’s communities, built by the improved unsupervised classification approach called the ‘Fuzzy K-mode’. These communities are also semantically modeled with modular profile ontology. To validate our proposed approach, we implemented a system for searching the itineraries for public transport. Finally, this thesis proposes research perspectives based on the limitations we encountered.

with which results. Through the analyses of data collected from Facebook's profiles of regional

Mobilitätseingeschränkte Fußgänger, wie etwa Rollstuhlfahrer, blinde und sehbehinderte Menschen oder Senioren, stellen besondere Anforderungen an die Berechnung geeigneter Routen. Die kürzeste Route ist nicht immer die am besten geeignete. In dieser Arbeit wird das Verfahren der multimodalen Annotation entwickelt, welches die Erweiterung der geographischen Basisdaten durch die Benutzer selbst erlaubt. Auf Basis der durch das Verfahren gewonnenen Daten werden Konzepte zu personalisierten Routenberechnung auf Grundlage der individuellen Anforderungen der Benutzer entwickelt. Das beschriebene Verfahren wurde erfolgreich mit insgesamt 35 Benutzern evaluiert und bildet somit die Grundlage für weiterführende Arbeiten in diesem Bereich. / Mobility impaired pedestrians such as wheelchair users, blind and visually impaired, or elderly people impose specific requirements upon the calculation of appropriate routes. The shortest path might not be the best. Within this thesis, the concept of multimodal annotation is developed. The concept allows for extension of the geographical base data by users. Further concepts are developed allowing for the application of the acquired data for the calculation of personalized routes based on the requirements of the individual user. The concept of multimodal annotation was successfully evaluated incorporating 35 users and may be used as the base for further research in the area.

Healthcare practices are changing as focus shifts fromtreating acute illnesses to chronic diseases. Theresponsibility of managing the treatment has shifted fromhealthcare providers to the individual in a higher degree. Toachieve good treatment the patients need to be empoweredso that they understand their condition and can makeinformed choices throughout their self-care. A researchthrough design approach was used to investigate how todesign a personalized empowering application for heartfailure patients. Aside from information relating to thecondition, the themes of physical activity, dieting and socialconnectedness were identified as central to address for theempowerment of this group. Patients, partners andhealthcare providers contributed with different perspectivesthroughout the design process. As a result five personas,representing potential users, were developed. Based on thepersonas and knowledge of the domain, user scenarios incurrent- and preferred state were constructed in order toguide the design of the empowering application called ‘TheHeart Companion’. It is a tablet application catering to thedifferent needs of the personas that also addresses the threethemes relevant for empowerment. The purpose of theapplication is to facilitate better understanding, a feeling ofsafety and a more active empowered life for the patient.The application enables personalization of the content byproviding bookmarking and addresses empowerment ofphysical activity by enabling various guided exercisesessions, personalized feedback, the possibility of reflectionand construction of personalized exercise sessions. / Sjukvårdens praxis förändras i takt med att fokus skiftar från behandling av akuta sjukdomar till kroniska sjukdomar. Ansvaret för behandling av kroniska sjukdomar har överförts från vårdgivare till att inbegripa individen till en högre grad. För att uppnå god behandling måste patienterna stärkas så att de förstår sitt tillstånd och kan fatta välgrundade beslut i sin egenvård. ’Research through design’ användes som metod för att undersöka hur en personifierad stärkande applikation för hjärtsviktspatienter kan utformas. Bortsett från information om tillståndet så identifierades fysisk aktivitet, kosthållning och social samhörighet som centrala teman att adressera för att stärka denna grupp. Patienter, partners och vårdgivare bidrog med olika perspektiv under designprocessen. Som ett resultat utav detta utvecklades fem personas för att representera potentiella användare. Baserat på personorna och kunskap om domänen så konstruerades användarscenarion i det nuvarande- samt det föredragna tillståndet för att vägleda designen av den stärkande applikationen som kallas Hjärtkompanjonen. Hjärtkompanjonen är en tablet-applikation som tillmötesgår de olika personornas behov och som även tar upp de tre teman som är relevanta att införliva för att stärka hjärtsviktspatienter. Syftet med applikationen är att åstadkomma bättre förståelse, en känsla av säkerhet och ett mer aktivt stärkt liv för patienten. Applikationen möjliggör personalisering av innehållet genom att tillhandahålla bokmärkning och adresserar hur en patient kan stärkas vid fysisk aktivitet genom att öppna upp för olika guidade träningspass, personlig feedback, möjligheter till reflektion samt konstruktion av personliga träningspass.

Background: As of today, it is apparent that with the ever-changing demands and needs of customers, companies are facing enormous pressure to deliver the right value, on time, in the right way, and proper manner. To realize the full potential of Artificial Intelligence (AI), a careful plan and method need to be established in the development and deployment when incorporating the technology with marketing. Technology is evolving at a rapid pace and Artificial Intelligence (AI) can be found in a variety of applications. AI in marketing can provide valuable data clusterization and insights for personalized recommendations, customer segmentation, or even advertising optimization.  Problem:  To date, a few studies have been made due to the rapid development of AI which has shown an opportunity for marketers. From this hype, companies are looking into speedy implementation where one can forget that this technology comes with risks and threats. “The problem is that everybody has unconscious biases and people embed their own biases into technology” (Kantayya, 2021). Although machines can deliver personalized numerical information, it cannot deliver new solutions such as products and services, nor classify different outputs with a cognitive mindset which could result in biased results. The objective of this research is to utilize the information and insights gathered from experts in the field of engineering and marketing to gain a holistic view of the current and future capabilities of AI in marketing.  Purpose: The focus of this bachelor thesis is to provide additional insights in regards to Artificial Intelligence in relation to marketing, taking into consideration bias, personalization, the black box, along with other possible implications of AI systems, also referred to as the dark side. To fulfill the researchers’ objective, qualitative interviews with practitioners and employees with different roles within the field of AI and Marketing were conducted. The paper will be focusing on concepts, theories, secondary data, and interviews which will be further discussed and give opportunities for future research.   Method: To perform this research, a qualitative research design was applied, and 12 structured interviews were conducted with those who have knowledge and experience with AI, marketing, or both.  Results: The study elucidates the potentials and fallbacks of Artificial Intelligence in marketing. Where the findings suggest a mixture of human intervention and technology is needed to work against the perceptions, bias, and manipulation the technology can possess. The aims then guide towards the conclusion presenting the important cognitive and emotional skills that humans obtain that are currently lacking in AI.  This study finds several key areas both in terms of opportunities and risks. Such key areas involve the possibility of delivering new, unique personalized content to a mass audience at lightning-quick speed and at the same time presenting a handful of risks by giving machines the permission to make human decisions. Risks found in this study presented as the dark side include the bubble, bias, manipulation, fear of losing jobs, lack of transparency creating the black-box phenomena. Therefore, this research is interesting especially for marketing managers in how AI could be used both from an opportunity perspective and possible risks to consider.

The data generated from the user's activities and surfing online can have a significant impact when used to improve the content of web pages. The user has little time to scan the content of a web page and the motivation to save time is high. Recommendations and personalized content are becoming more and more common and both businesses and consumers seem to benefit from more targeted content. In order to know what recommendations to present to each individual, information about the user is needed and there are many tools to analyze the behaviours of the users online. In this study, we examined if relevant content on the front-page of Visma, a business software provider, can increase the interaction on the front-page and what metrics can be used to measure the interaction. Together with experts in the field we investigated the eAccounting software users’ preferences for targeted recommendations. By having a pop up survey for users of Visma, conducting an A/B test and using a heatmap for the eAccounting users we found out that a more customized front-page makes the user interact more with the page.

Background The cosmetic industry is growing at a fast pace and, like other industries, there are a few problems regarding e-retailing that makes it hard for customers to make optimal purchase decision. Example of problems are; the number of e-retailers increase, e-retailers’ variety of products increase, and the content is almost unlimited. Companies have started to personalize to solve these problems. However, there is a lack in the literature regarding how-to best design personalization with regard to its dimensions and subcategories. Therefore, the next step is to get an understanding of which dimensions and subcategories that are the most effective. Objectives At present, the majority of the literature regarding personalization focus on how personalization can be performed and not in which way personalization should be performed to increase customer satisfaction and customer loyalty. This means that the knowledge regarding the subcategories within personalization and their effectiveness is limited. Therefore, the purpose of this thesis is to increase the body of knowledge regarding which subcategories that is the most effective in increasing customer satisfaction and/or loyalty. Knowledge about this topic can give companies perspectives to know how personalization should be designed to increase customer satisfaction and customer loyalty. Methods This study was a survey type, resulting in primary quantitative data collecting and analyzes. A pilot test was first made, before the final questionnaire. This to test the reliability and validity. The data from the pilot study were partly analyzed in SPSS and partly analyzed manually. The data from the final questionnaire were then analyzed in SPSS. Results Regarding customer satisfaction, there is a statistically significant difference between content personalization and AIGUI (Adaptive Interface and Graphical User Interface) personalization as well as between AIGUI personalization and combination (content and AIGUI) personalization. However, there is no statistical difference between content personalization and combination personalization. Regarding customer loyalty, the is no statistical difference between the personalization groups. Conclusions An e-retailer in the cosmetic industry should focus on content personalization or combination personalization / Bakgrund Industrin om kosmetik växer i en snabb takt och likt andra industrier finns det problem, kring återförsäljare som använder sig utav e-handel, som gör det svårt för kunderna att göra optimala köpbeslut. Exempel på problem är; Antalet återförsäljare som använder sig utav e-handel ökar, deras variationer av produkter ökar och informationen är i princip oändlig. Mål Litteraturen kring personligfiering handlar till större delen om hur personligfiering kan utformas och inte på vilket sätt personligfiering borde utformas för att öka kundnöjdhet och kundlojalitet. Detta innebär att kunskapen kring underkategorier av personligfiering och dess effektivitet är begränsad. Därför är syftet med studien att öka kunskapen kring underkategorier av personligfiering och ta reda på hur dessa påverkar kundnöjdhet och/eller kundlojalitet. Kunskap om detta kan ge företag perspektiv på hur de borde utforma sin personligfiering för att öka kundnöjdhet och kundlojalitet. Metod Denna studien använde sig utav en undersökning för att besvara avhandlingsfrågan. Detta resulterade i en studie av primär kvantitativ datainsamling och analysering. Först genomfördes en pilotstudie för att testa pålitligheten och giltigheten. Datan från pilotstudien var delvis analyserad i SPSS och delvis analyserad manuellt. Datan från den slutgiltiga käten analyserades i SPSS. Resultat Vad gäller kundnöjdhet framkom det att det finns en statistisk signifikant skillnad mellan innehålls personligfiering och AIGUI (Adaptive Interface and Graphical User Interface) personligfiering. Det finns också en statistisk signifikant skillnad mellan AIGUI personligfiering och combination (innehåll och AIGUI) personligfiering. Däremot framkom det att det inte finns någon statistisk signifikant skillnad mellan innehålls personligfiering och combination personligfiering. Vad gäller kundlojalitet framkom det att det inte finns någon statistisk signifikant skillnad mellan de olika personligfieringstyperna. Slutsatser En återförsäljare av kosmetik som använder sig utav e-handel borde fokusera på innehåll personligfiering eller combination personligfiering.

As society progresses towards a more algorithmic era, the influence of artificial intelligence (AI) is driving a revolution in the digital landscape. At its core, AI applications aim to engage customers by providing carefully tailored and data-driven personalization and customization of products, services, and marketing mix elements. However, the adoption of AI, while promising enhanced personalization, poses challenges due to the increased collection, analysis, and control of consumer data by technology owners. Consequently, concerns over data privacy have emerged as a primary consideration for individuals. This paper delves deeper into the implications of the personalization- privacy paradox, aiming to provide a comprehensive analysis of the challenges and opportunities it presents. The purpose of this thesis is to understand users’ privacy concerns and willingness to disclose their personal information to AI technologies by addressing the limitations of previous research and utilizing qualitative methods to gain a more in-depth understanding of consumer views. To understand users’ privacy concerns and willingness to disclose personal information to AI technologies, a qualitative approach was followed. Combining a deductive and inductive approach to fulfill the purpose of the study, empirical data was collected through 20 semi- structured interviews. The participants were chosen using a purposive sampling technique. Users’ privacy concerns and willingness to disclose personal information to AI technologies differ significantly. It depends not only on the individual, but also on the type of AI technology, the company providing the AI technology, the possibility of obtaining additional benefits, and whether the company is transparent about its data collection and can provide proof of security.

Effective data visualization is essential for digital self-tracking to help users gain insights into their behavior and habits. Personalized visualizations engage users, making the self-tracking experience more meaningful. However, potential biases and limitations should be considered to ensure an accurate and objective self-tracking process. The study aims to examine users' interactions with visualized data in the digital self-tracking process and understand their perceptions of the accuracy and objectivity of personal data visualizations and the self-tracking processes on platforms that offer self-tracking features. These platforms include applications for tracking health and fitness, habits, music listening, book reading, and movie watching. The study employs a qualitative approach using semi-structured interviews with an ethnographic approach as a data collection method. This approach was selected to investigate users' interactions and opinions on data visualizations and the digital self-tracking process.  The findings show that participants primarily use data visualizations and other personal visualizations as reminders, for comparisons, planning, and motivation. Although they do not extensively analyze the visualized data, participants report experiencing heightened self-awareness and motivation. Despite their awareness of potential inaccuracies and subjectivity in the visualizations and the self-tracking process, participants are willing to overlook these aspects due to the perceived benefits associated with the process. Moreover, participants generally express a level of trust in the accuracy of their visualized data.

Remote electrical tilt (RET) optimization involves maximizing the coverage and minimizing interference for antennas in a cellular network. A RET optimization problem typically has many of antennas, each of which has little data. Reinforcement learning (RL) agents have recently been deployed to solve RET optimization problems [1, 2]. These algorithms generally require large amounts of data, and therefore they are not applied for individual antennas but rather for groups of antennas. We show that this leads to degraded performance than agents personalized for individual antennas with extensive data. Furthermore, we design a reward shaping (RS) agent, which augments the reward signal to learn quicker than agents trained only on individual antennas while still retaining their performance. / Remote electrical tilt (RET) optimering innebär att försöka maximera täckningen och minimera störningar för antenner i ett mobilnät. Ett RET-optimeringsproblem har vanligtvis ett stort antal antenner, som var och en har lite data. Reinforcement learning (RL)-agenter har nyligen använts för att lösa RET-optimeringsproblem [1, 2]. Dessa algoritmer kräver i allmänhet stora mängder data och därför används de inte för enskilda antenner utan snarare för grupper av antenner. Vi visar att detta leder till försämrad prestanda jämfört med agenter anpassade för individuella antenner med stora datamängder. Dessutom designar vi en Reward shaping (RS)-agent, som förstärker belöningssignalen för att lära sig snabbare än agenter som bara tränas på individuella antenner samtidigt som deras behåller sin prestanda.