Les réseaux sociaux ont connu une évolution dramatique et ont été utilisés comme des moyens pour exercer plusieurs activités. En fait, via les réseaux sociaux, les utilisateurs peuvent découvrir, gérer et partager leurs expériences et avis en ligne. Cependant, la nature ouverte et décentralisée des réseaux sociaux les rend vulnérables à l'apparition des utilisateurs malveillants. Par conséquent, les utilisateurs éventuels peuvent faire face à plusieurs de problèmes liés à la confiance. Ainsi, une évaluation de confiance effective et efficace est très importante pour la prise de décisions par ces utilisateurs. En effet, elle leur fournit des informations précieuses leur permettant de faire la différence entre ceux dignes et indignes de confiance. Cette thèse a pour but de fournir des méthodes de gestion de confiance et de réputation des utilisateurs des réseaux sociaux efficaces et qui peuvent être présentées par les quatre contributions suivantes. La première contribution présente une complexe extraction des contextes et des intérêts des utilisateurs, où les informations contextuelles sociales complexes sont prises en compte, reflétant mieux les réseaux sociaux. De plus, nous proposons un enrichissement de l'ontologie Dbpedia par des concepts de folksonomies.Ensuite, nous proposons une approche de gestion de la confiance, intitulée IRIS, permettant la génération du réseau de confiance et le calcul de la confiance directe. Cette approche considère les activités sociales des utilisateurs incluant leurs relations sociales, préférences et interactions.La troisième contribution de cette thèse est la gestion de transitivité de confiance dans les réseaux sociaux. En fait, c'est nécessaire et significatif d'évaluer la confiance entre deux participants n’ayant pas des interactions directes. Nous proposons ainsi, un modèle d'inférence de confiance, appelé TISON, pour évaluer la confiance indirecte dans les réseaux sociaux.La quatrième contribution de cette thèse consiste à gérer la réputation des utilisateurs des réseaux sociaux. Pour ce faire, nous proposons deux nouveaux algorithmes. Nous présentons un nouvel algorithme exclusif pour la classification des utilisateurs basés sur leurs réputations, appelé le RePC. De plus, nous proposons un deuxième algorithme, FCR, qui présente une extension floue de RePC. Pour les approches proposées, nous avons conduits différentes expérimentations sur des ensembles de données réels ou aléatoires. Les résultats expérimentaux ont démontré que nos algorithmes proposés produisent de meilleurs résultats, en termes de qualité des résultats livrés et d’efficacité, par rapport à différentes approches introduites dans littérature / Online Social Networks (OSNs) have known a dramatic increase and they have been used as means for a rich variety of activities. In fact, within OSNs, usersare able to discover, extend, manage, and leverage their experiences and opinionsonline. However, the open and decentralized nature of the OSNs makes themvulnerable to the appearance of malicious users. Therefore, prospective users facemany problems related to trust. Thus, effective and efficient trust evaluation isvery crucial for users’ decision-making. It provides valuable information to OSNsusers, enabling them to make difference between trustworthy and untrustworthyones. This thesis aims to provide effective and efficient trust and reputationmanagement methods to evaluate trust and reputation of OSNs users, which canbe divided into the following four contributions.The first contribution presents a complex trust-oriented users’ contexts andinterests extraction, where the complex social contextual information is taken intoaccount in modelling, better reflecting the social networks in reality. In addition,we propose an enrichment of the Dbpedia ontology from conceptualizations offolksonomies.We second propose the IRIS (Interactions, Relationship types and Interest Similarity)trust management approach allowing the generation of the trust networkand the computation of direct trust. This model considers social activities of usersincluding their social relationships, preferences and interactions. The intentionhere is to form a solid basis for the reputation and indirect trust models.The third contribution of this thesis is trust inference in OSNs. In fact, it isnecessary and significant to evaluate the trust between two participants whomhave not direct interactions. We propose a trust inference model called TISON(Trust Inference in Social Networks) to evaluate Trust Inference within OSNs.The fourth contribution of this thesis consists on the reputation managementin OSNs. To manage reputation, we proposed two new algorithms. We introducea new exclusive algorithm for clustering users based on reputation, called RepC,based on trust network. In addition, we propose a second algorithm, FCR, whichis a fuzzy extension of RepC.For the proposed approaches, extensive experiments have been conducted onreal or random datasets. The experimental results have demonstrated that ourproposed algorithms generate better results, in terms of the utility of delivered results and efficiency, than do the pioneering approaches of the literature

No description available.

Vehicle automation has garnered increasing attention as a means of improving safety and efficiency. Advanced Driver Assistance Systems (ADAS) have gained popularity in the transport industry. However, establishing an appropriate level of trust in these systems is crucial for their successful implementation. This research explores the factors influencing driver trust calibration in different levels of automation within driver assistance systems for commercial mobility trucks to ensure drivers comprehend the limitations of these systems and uphold road safety. A qualitative approach involved eleven interviews and observations with drivers to explore their perceptions, experiences, and expectations regarding these systems. The study’s findings extend the Hoff and Bashir Trust model to include significant social factors in calibrating trust. These findings offer valuable insights into the various trust factors that impact driver trust calibration at different levels of automation in driver assistance systems for commercial mobility trucks. These insights contribute to academia in that they help understand how trust in automation is formed and calibrated in real-world settings. In the automotive industry, they can guide the design and implementation of these systems to enhance future drivers’ safety and overall experience.

In today's world, network security is of utmost importance for companies and authorities as data and information are increasingly being stored and transmitted through cloud solutions. This has led to a higher risk of cyberattacks. To enhance security during this transition, Zero Trust has emerged as a promising concept, with its direction from an internal IT-environment to a more secure cloud solution. Its main motto "never trust, always verify" emphasizes the importance of a high level of security. This study aims to highlight the significance of Zero Trust and compare the attitudes of Swedish companies and authorities towards its implementation. The study was conducted using semi-structured interviews with IT experts who possess knowledge in this domain. Additionally, a literature review was conducted to connect the outcomes of the interviews with the existing research in the field of Zero Trust. The results indicate that several barriers must be considered during the implementation phase. These include cost-related issues, lack of expertise, and outdated systems that cannot handle Zero Trust at the required security level. Despite these challenges, the respondents' attitudes towards Zero Trust were positive. They believed that it could be applied to sporadic and isolated parts of the network infrastructure, and that striving towards it at a theoretical level was essential.

Philosophiae Doctor - PhD / This dissertation explored the degree and distribution of social capital forms-social networks, volunteer associations, generalized trust, particularized trust and norms of reciprocity in rural Ethiopia. It aims to contribute to the academic understanding of social capital formation in Africa. Whereas the extent of literature focused on uncovering difference in the distribution of social capital at macro, meso and micro levels, this dissertation instead unpacks how political and economic conditions in rural Ethiopia shape the extent of the different forms of social capital at different levels. Aresi and Menze are selected for the case-study based on their representation of rural Ethiopia's ecological, language, and ethnicity, historical and religious diversity. A household survey (n = 735) was conducted to uncover the distribution of social capital forms at the individual, regional and aggregate levels. Eight focus group discussions and ten key informant interviews were conducted to unpack why the different forms of social capital unfold the way they do. The qualitative interviews were also used for validating the result of the household survey by uncovering how institutional and structural factors reconstituted and redefined at the local level and translated into social capital formation. The empirical result revealed that social networks and norms of reciprocity are the most important social capital apects in the study areas. At the same time, trust relations and formal networks (volunteer organizations) are the least important aspects of social capital. However, over time, community-organization has become an alternative venue where community members relate and support each other. Thus, network-based and community-based social capital are the principal social capital forms in the study areas. This result is largely inconsistent with the previous studies that have been conducted in developed and democratic countries which have asserted generalized trust and impersonal civic organizations as the major forms of social capital, rather than localized and interpersonal forms of social capital. The fact that different forms of social capital are present in different contexts with different levels of importance shows that forms of social capital do not necessarily articulate together. Moreover, the variation in the relative importance of social capital forms shows that the production of social capital is shaped by the prevailing political and economic conditions. Thus, this dissertation makes the case that it is necessary to distinguish among the different forms of social capital both in theory and practice.

No description available.

Automated systems are integral in the development of modern aircraft, especially for complex military aircraft. Pilot Trust in Automation (TIA) in these systems is vital for optimizing the pilot-vehicle interface and ensuring pilots use the systems appropriately to complete required tasks. The objective of this research was to develop and validate a TIA scale and survey methodology to identify and mitigate trust deficiencies with automated systems for use in Army Aviation testing. There is currently no standard TIA assessment methodology for U.S. Army aviation pilots that identifies trust deficiencies and potential mitigations. A comprehensive literature review was conducted to identify prominent TIA factors present in similar studies. The compiled list of factors and associated definitions were used in a validation study that utilized the Analytic Hierarchy Process (AHP) as a pair-wise comparison tool to identify TIA factors most relevant to Army pilots. A notional survey, the Aviation Systems – Trust Survey (AS-TS), was developed from the identified factors and pilots were used as subjects in scenario-based testing to establish construct validity for the survey. Exploratory factor analysis was conducted after data collection and a validated survey was produced. A follow-on study interviewed Army test and evaluation experts to refine the survey methodology and ensure appropriate context for the recommended mitigations. A final packet was developed that included instructions for the rating scale, associated item definitions, and recommended mitigations for trust deficiencies. Future research will focus on other Army demographics to determine the generalizability of the AS-TS.

Automated trust negotiation is the process of establishing trust between entities with no prior relationship through the iterative disclosure of digital credentials. One approach to negotiating trust is for the participants to exchange access control policies to inform each other of the requirements for establishing trust. When a policy is received at runtime, a compliance checker determines which credentials satisfy the policy so they can be disclosed. In situations where several sets of credentials satisfy a policy and some of the credentials are sensitive, a compliance checker that generates all the sets is necessary to insure that the negotiation succeeds whenever possible. Compliance checkers designed for trust management do not usually generate all the satisfying sets. In this thesis, we present two practical algorithms for generating all satisfying sets given a compliance checker that generates only one set. The ability to generate all of the combinations provides greater flexibility in how the system or user establishes trust. For example, the least sensitive credential combination could be disclosed first. These ideas have been implemented in TrustBuilder, our prototype system for trust negotiation.

Infidelity is one of the most prevalent presenting problems in relational therapy. There are many conceptualizations of the healing process following infidelity and suggested interventions and treatment plans. Forgiveness is an essential part of relationship growth and healing interpersonal hurts. Reconciliation is relationship repair that can accompany forgiveness. In addition, restoring trust is essential to reconciliation. Means to accomplish rebuilding trust must be established. This study seeks to expand understanding of the treatment of infidelity specifically on the topics of forgiveness, reconciliation, and restoration of trust through disclosure. The Forgiveness vs. Reconciliation and Trust Scale (FvRTS) was developed as a measure for therapists' views and perception concerning these three topics. Through statistical analysis of the FvRTS, which was administered to relational therapists, this study investigates the views and opinions held by therapists concerning the relationship between reconciliation and forgiveness and the role of disclosure, both initial and ongoing, in restoring trust following infidelity. Therapists conveyed the view that reconciliation is not required for forgiveness. However, in their practice they encourage reconciliation following an affair. Results indicated that therapists perceive initial disclosure as having immediate negative impacts. But overall, therapists expressed a strong view that both initial and ongoing disclosure have a positive long term impact on relational healing.

BAKGRUND: Den automatiserade ekonomiska rådgivningen i form av robotrådgivning är ett nytt fenomen som funnits på den svenska marknaden i knappt 15 år. Den tidigare populära traditionella rådgivningen har i takt med robotrådgivningens framfart fått ett konkurrenskraftigt alternativ. Förtroende anses vara en av de viktigaste aspekterna för individen när det kommer till valet av ekonomisk rådgivare, men trots att flera studier gjorts inom förtroendeområdet vid ekonomisk rådgivning saknas det fortfarande kunskap om vilka faktorer som förklarar förtroendet till robotrådgivning. Således finns incitament för att täcka den kunskapslucka som idag finns gällande individer som ej aktivt använder sig av robotrådgivare.    SYFTE: Syftet med studien är att skapa förståelse kring vilka faktorer som förklarar svenska individers förtroende till robotrådgivning utifrån ett icke-användande perspektiv, samt hur dessa faktorer främjar eller hämmar förtroendet till tjänsten.   GENOMFÖRANDE: Studien har genomförts med en kvalitativ metod genom semi-strukturerade intervjuer med individer som ej aktivt använder robotrådgivning men som ändå besitter kunskap om hur tjänsten fungerar. Respondenterna fick genom intervjuerna besvara fördjupande frågor om specifika förtroendefaktorer samt hur de uppfattar förtroendet till robotrådgivning.   SLUTSATS: Studiens resultat utifrån den insamlade empirin visade att respondenterna har ett högt förtroende till robotrådgivning där kompetens och följdriktighet visade sig vara de främst främjande faktorerna till förtroendet. Kommunikation var delvis främjande, delvis hämmande, medan medkänslan i och med sin avsaknad vid robot-rådgivning framhävdes som en svagt främjande faktor. Anseende, bemötande, kostnad och transparens lyftes också som förklarande förtroendefaktorer till robotrådgivning. / BACKGROUND: The automated financial advisor in the form of robo-advisors is a new phenomenon that has existed in the Swedish market for almost 15 years. The already popular traditional financial advisory has, in step with the progress of robo-advisory, been given a competitive alternative. Trust is considered as one of the most important aspects for the individual when it comes to choosing financial advisors, but despite several studies done in the area of ​​trust in financial advisory, there is still a lack of knowledge about which factors are explaining trust in robo-advisory. Thus, there are incentives to cover the knowledge gap that currently exists for individuals who do not actively use robo-advisory.   AIM: The aim with this thesis is to gain knowledge about which factors are explaining individuals’ trust in robo-advisors from a non-user perspective, as well as to investigate in what way these factors might enable or disable the trust in the service.    COMPLETION: To fulfill the aim of this study, a qualitative method has been implemented through semi-structured interviews with individuals that do not actively use a robo-advisor. Through the interviews, the respondents were asked to answer in-depth questions about specific trust factors and how they perceive the trust in robo-advisory.   CONCLUSION: The overall results of the study based on the collected empirical data showed that the respondents have a high level of trust in robo-advisory, where competence and consistency proved to be the main enabling factors for trust. Communication was partly enabling, partly disabling, while compassion, due to its lack within robo-advisory, was emphasized as a vague promoting factor. Reputation, treatment, cost and transparency were also highlighted as explanatory factors of trust in robo-advisory.