Ionian University
Από τη μελέτη συμπεριφοράς και προφίλ χρηστών κινητών εφαρμογών στην εκπαίδευση: προσεγγίσεις για την ενίσχυση της ενημερότητας πληροφοριακής ιδιωτικότητας
Abstract
dc:descriptionSmartphones have become an indispensable tool, supporting activities in almost every aspect of our daily lives. However, alongside the capabilities they offer, they collect vast amounts of personal data, putting users' privacy at risk. Many mobile applications access sensitive user data not only to provide their core functionality but also for advertising and other purposes. Access rights to most of this data, such as phonebook contacts, location, camera, photos, microphone, and more, are granted by the device users themselves through application permission requests.Considering the multitude of applications modern users install on their mobile devices, it becomes evident that they face numerous and complex decisions regarding their information privacy, the volume of which can quickly become unmanageable. As highlighted in a study by Smullen et al. (2020), the average Android user must make more than a hundred privacy decisions just to configure app-related permission settings. Furthermore, Raber & Krueger (2017) report that an average user with 95 installed applications faces over 400 permission-granting decisions.Therefore, it comes as no surprise that the majority of users do not dedicate time to configuring many of these settings. In fact, some become accustomed to such requests and respond automatically (habituation) without realizing the consequences of their choices. As a result, few individuals actually read the required permissions, and even fewer understand them correctly. At the same time, a significant number of users express surprise and dismay when they realize the extent to which sensitive data is collected and utilized by applications. Certain users are not sufficiently aware of the risks, whereas others wish to protect their data but lack the requisite knowledge and skills. Finally, there are those who, despite being aware of the risks, fail to take practical measures and continue to disclose their data. Consequently, in all scenarios, users remain deeply vulnerable, essentially constituting the weak link in the protection of their personal data. As privacy management becomes an increasingly complex process for the average user, the need for their substantial support emerges as an imperative. The research focus of the present dissertation, titled 'From studying mobile application users behaviors and profiles to training: Approaches towards enhancing information privacy awareness' is to provide support to users in managing their privacy within applications, as well as on the Internet in general, along a dual axis: at a technological level, through the use of automated machine learning tools, and at a cognitive level, through targeted education and training. More specifically, the objective of this study is to contribute to the enhancement of users' information privacy through the development of a commonly accepted framework - a Common Body of Knowledge (CBK) - to serve as the foundation for their education and training. Concurrently, it employs machine learning techniques in an effort to further comprehend the factors that influence their privacy decisions across various applications. The introductory section of this dissertation provides an in-depth description of the challenges faced by modern online users regarding the management of their information privacy, specifically within the context of mobile applications and their permission requests. Consequently, it highlights the imperative need for user support through both the utilization of automated machine learning tools and the provision of appropriate education and training. Furthermore, a significant gap in the literature is identified concerning the lack of a framework that could serve as a foundation for designing information privacy awareness and training programs, as well as for organizing relevant educational materials. The present dissertation poses four research questions aligned with this objective.To address the first research question 'What must a user know in order to make informed privacy protection decisions when managing application permissions?' the permission ecosystem of the Android operating system was examined, bridging the theoretical framework with empirical analysis. Initially, the model's architecture, the various types of permissions, and the mechanism for managing such requests were analyzed. Subsequently, an empirical study was conducted involving the retrieval, categorization, and analysis of so-called 'dangerous permissions' from seven selected popular applications, with the aim of assessing the resulting privacy risks to users. This process leads to valuable conclusions and the proposal of practical advice for user protection, alongside the design of relevant educational workshops aimed at further raising public awareness and enhancing knowledge on information privacy issues. To investigate the second research question, 'Which factors shape users' privacy concerns and self-reported behaviors regarding the management of application permissions?', as well as the third question, 'Which factors influence users' decisions to accept or reject application permissions?', an extensive literature review was conducted concerning the factors that impact users' privacy behaviors and their decisions regarding application requests for access to device resources. Through this systematic literature review, the factors influencing individuals' privacy decisions were compiled, and a structured overview is provided in a tabular format, classifying them into two distinct categories: application-related factors and user-centric factors.Subsequently, the study examines how the existing literature utilizes privacy profiles as predictive tools for users' choices regarding application permissions. Furthermore, it analyzes the utilization of Machine Learning techniques and other Artificial Intelligence methods, as documented in the literature, for the reliable prediction of users' privacy preferences and decisions. This approach serves as an effective mechanism to alleviate the burden placed on users by continuous decision-making. This comprehensive literature review culminates in the development of a novel research model, aiming to further investigate the factors that impact users' privacy behaviors and decisions concerning application permissions. In the proposed research model, we integrate traditional application-related factors (e.g., application category, permission type) with under-explored user-centric variables, in an endeavor to extend the existing literature.More specifically, we examine the privacy practices adopted by users during the installation of an application, as well as those related to the management of permissions and privacy settings. Concurrently, the technical knowledge and skills of the users are investigated, both from their own perspective (self-reported data) and in terms of their theoretical knowledge regarding application permissions (objective perspective), a topic that has been insufficiently addressed in the literature. Furthermore, we explore users' comprehension of application permissions, their information privacy concerns, and their comfort levels regarding the sharing of personal data. To evaluate the aforementioned factors, we developed a closed-ended questionnaire and conducted a survey involving 60 volunteer Android device users who were asked to complete it. Subsequently, we extracted actual data from the mobile devices of the survey participants; more specifically, we retrieved the permissions they had granted to their installed applications. Following this, we present the research results along two axes. Initially, we analyze the quantitative data derived from the participants' questionnaire responses, examining the impact of demographic characteristics and the correlations among the research variables. Additionally, we categorize the users into two distinct privacy profiles by applying machine learning algorithms to the collected data. Thus, the different behavioral patterns of the participants regarding the management of their personal data are highlighted, along with the significance of technical knowledge in translating privacy concerns into protective actions. Proceeding to the next stage of the research, we contrast self-reported behaviors, as captured in the questionnaires, with users' actual behavior, as recorded by their devices. The creation of the Relative Permission Rejection Index (PRI) plays a pivotal role in this process, objectively evaluating their protective practices in the context of applications and permission requests. Through this integrative analysis, we draw valuable conclusions regarding how users perceive and actively protect their personal data. Finally, we extend our study by employing Machine Learning algorithms, developing a decision tree that delineates the factors decisively influencing users' ultimate acceptance or rejection of application permissions. The research findings demonstrate that gaps in privacy protection do not stem from user apathy, but rather reflect a tangible 'technical skills gap'. It is therefore concluded that mere theoretical awareness is insufficient to alter user behavior. To transform passive concern into proactive behavior, the technical empowerment of users and their systematic training in the utilization of practical protection tools are absolutely essential. This conclusion was further corroborated by the Machine Learning models, which highlighted that proactive privacy protection is predominantly a byproduct of digital literacy. Key pillars in this regard include the systematic management of permissions and privacy settings (PMP), the understanding of access requests (UAP), and maintaining a critical stance during the installation of new applications (AIP). Consequently, the targeted education of users in the utilization of practical privacy management tools emerges as an imperative need. Guided by these comprehensive findings, the present study formulates specific proposals aimed at both optimizing user privacy protection and charting future research directions. The approach to the fourth research question 'What must a modern user know, on both a theoretical and practical level, to be capable of facing contemporary challenges while simultaneously protecting their information privacy?' was formulated following an exhaustive review of the literature surrounding information privacy. Specifically, this review focused on online users' perceptions, contemporary threats, and protection strategies. Concurrently, it examined the legal and regulatory frameworks for information privacy protection, organizational practices regarding consumer data management, and the critical importance of users' digital awareness and education concerning the protection of their privacy. Thus, the theoretical foundation was established for the creation of a Common Body of Knowledge (CBK) regarding information privacy, addressing a domain where no comparable research endeavor has been previously documented. Following the conceptual development of the proposed InfoPrivacy CBK, it was depicted through concept maps. Subsequently, empirical research was conducted for its validation and refinement, involving the participation of seven domain experts. To facilitate its evaluation, an electronic questionnaire comprising both structured and open-ended questions was developed. The experts' evaluation was exceptionally positive, confirming the soundness of the framework's structure and content, as well as the successful fulfillment of its educational objectives. Ultimately, this yielded an original and innovative CBK, which effectively bridges a significant gap in the existing literature.Within the scope of this dissertation, a comprehensive conceptual and methodological framework is developed through the InfoPrivacy CBK, aimed at supporting the design of information privacy awareness and training programs, as well as the organization of relevant educational material. Our research makes a multidimensional contribution, extending across theoretical, methodological, and practical levels. Specifically, the theoretical contribution of the dissertation encompasses the following points: •Formulation of a Common Body of Knowledge (CBK) for Information Privacy: The creation of an innovative framework for information privacy, which addresses a significant literature gap and serves as a central pillar for the design of modern education and training programs. •Foundation for educational interventions: Enhancing the literature by providing the necessary theoretical background for designing initiatives that combine theory and practical application regarding information privacy, with a particular focus on application permissions. •Development of a novel research model and behavioral mapping: Enriching the literature with an extended predictive model that decodes users' decision-making processes regarding app permissions. The model innovates by integrating traditional application characteristics with under-explored user-centric variables, such as behavior during app installation, management of permissions and privacy settings, permission comprehension, privacy concerns, comfort levels, as well as users' technical knowledge and skills. •Theoretical substantiation of digital literacy: Through the investigation of the aforementioned factors, the decisive role of technical empowerment and digital literacy in the average user's transition from passive concern to proactive privacy protection is highlighted and theoretically substantiated, concurrently providing a strong impetus for future research.At a methodological level, the contribution of the dissertation encompasses the following points: •Structuring a framework for the design of educational interventions: The provision of a fully structured and theoretically grounded framework, which follows a clear sequence of steps for the design and creation of constructive educational actions regarding information privacy and application permissions. •Methodology for practical implementation: The formulation of applicable methodological approaches for the practical implementation of educational and training interventions, centered on the mobile device ecosystem and access permission mechanisms. •Innovation in the quantification of digital behavior (PRI Index): The introduction and standardization of the Relative Permission Rejection Index (PRI). This constitutes a novel, composite metric tool that transcends traditional approaches, facilitating a more effective and comprehensive quantification of users' actual digital behavior by directly factoring in the operational context of each application. •Development of novel research tools (Questionnaires): The design and provision of two specialized metric tools for data collection: a questionnaire for the evaluation of curricula by instructors and learners, alongside a targeted tool for measuring specific aspects of information privacy with an emphasis on app permission decisions.Finally, the contribution of the dissertation at a practical level encompasses the following points: •Utilization of the InfoPrivacy CBK as a 'roadmap': The proposed framework serves as a practical tool for educational institutions, privacy awareness program designers, information security software companies, organizations, corporations, and education providers, guiding the design of targeted educational interventions and comprehensive awareness programs regarding information privacy. •Strategic guidance for policymaking: The provision of evidence-based guidelines to relevant authorities, enabling them to focus on modern and efficient methods and means for the broad enhancement of citizens' digital awareness. •Substantial digital empowerment of users: Through their participation in theoretically grounded education and training programs, users acquire the necessary tools to sharpen their critical thinking and protect their privacy by making fully informed decisions. This individual empowerment will gradually lead to the creation of digital communities characterized by heightened awareness and collective sensitivity towards information privacy. •Optimization of software design: The provision of actionable guidelines for developers, software engineers, and platform designers, aiming at the development of more transparent and efficient access permission mechanisms, thereby enabling users to make more informed decisions regarding their privacy.
Degree
thesis:*- Grantor dc:publisher
- Ionian University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Lavranou, Eirini
- Λαβράνου, Ειρήνη
Subjects
dc:subject × 26- Ενημερότητα πληροφοριακής ιδιωτικότητας
- Εκπαίδευση χρηστών
- Εφαρμογές κινητών συσκευών
- Άδειες εφαρμογών
- Προφίλ ιδιωτικότητας χρηστών
- Κοινό Σύνολο Γνώσης
- Ανάλυση συμπεριφοράς
- Information privacy awareness
- User training
- Mobile applications
- Application permissions
- User privacy profiles
- Common Body of Knowledge (CBK)
- Behavior analysis
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Φυσικές Επιστήμες
- Διάδραση ανθρώπου-υπολογιστή
- Εκπαίδευση
- Κοινωνικές Επιστήμες
- Εκπαίδευση, άλλοι τομείς
- Computer and Information Sciences
- Natural Sciences
- Human-Computer Interaction
- Education
- Social Sciences
- Education, miscellaneous
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/61823