Search for: All records

Lack of diversity and high dropout rates among underrepresented students plague the CS discipline. We developed, administered, and validated survey scales measuring social factors that impact the retention and graduation of under-represented CS undergrads at two institutions. Results revealed significant differences between students who identify as men vs. women in terms of computing identity and confidence, and between black and non-black students in terms of familiarity with future opportunities. 

This position paper describes our research project to improve middle school students’ use of security “best-practices” in their day-to-day online activities, while enhancing their fundamental understanding of the underlying security principles and math concepts that drive AI and cybersecurity technologies. The project involves the design and implementation of a time- and teacher-friendly learning module that can be readily integrated into existing middle school math curricula. We plan to deploy this module at a high-needs, rural-identifying middle school in South Carolina that serves underrepresented students 

The prevalence of smartphones in our society warrants more research on understanding the characteristics of users and their information privacy behaviors when using mobile apps. This paper investigates the antecedents and consequences of “power use” (i.e., the competence and desire to use technology to its fullest) in the context of informational privacy. In a study with 380 Android users, we examined how gender and users’ education level influence power use, how power use affects users’ intention to install apps and share information with them versus their actual privacy behaviors (i.e., based on the number of apps installed and the total number of “dangerous permission” requests granted to those apps). Our findings revealed an inconsistency in the effect of power use on users’ information privacy behaviors: While the intention to install apps and to share information with them increased with power use, the actual number of installed apps and dangerous permissions ultimately granted decreased with power use. In other words, although the self-reported intentions suggested the opposite, people who scored higher on the power use scale seemed to be more prudent about their informational privacy than people who scored lower on the power use scale. We discuss the implications of this inconsistency and make recommendations for reconciling smartphone users’ informational privacy intentions and behaviors. 

To account for privacy perceptions and preferences in user models and develop personalized privacy systems, we need to understand how users make privacy decisions in various contexts. Existing studies of privacy perceptions and behavior focus on overall tendencies toward privacy, but few have examined the context-specific factors in privacy decision making. We conducted a survey on Mechanical Turk (N=401) based on the theory of planned behavior (TPB) to measure the way users’ perceptions of privacy factors and intent to disclose information are affected by three situational factors embodied hypothetical scenarios: information type, recipients’ role, and trust source. Results showed a positive relationship between subjective norms and perceived behavioral control, and between each of these and situational privacy attitude; all three constructs are significantly positively associated with intent to disclose. These findings also suggest that, situational factors predict participants’ privacy decisions through their influence on the TPB constructs. 

This report documents the program and the outcomes of Dagstuhl Seminar 23031 Frontiers of Information Access Experimentation for Research and Education, which brought together 38 participants from 12 countries. The seminar addressed technology-enhanced information access (information retrieval, recommender systems, natural language processing) and specifically focused on developing more responsible experimental practices leading to more valid results, both for research as well as for scientific education.

The seminar featured a series of long and short talks delivered by participants, who helped in setting a common ground and in letting emerge topics of interest to be explored as the main output of the seminar. This led to the definition of five groups which investigated challenges, opportunities, and next steps in the following areas:reality check, i.e. conducting real-world studies, human-machine-collaborative relevance judgment frameworks, overcoming methodological challenges in information retrieval and recommender systems through awareness and education, results-blind reviewing, and guidance for authors.

Date:15--20 January 2023.

Website:https://www.dagstuhl.de/23031.

 

Fitness trackers are undoubtedly gaining in popularity. As fitness-related data are persistently captured, stored, and processed by these devices, the need to ensure users’ privacy is becoming increasingly urgent. In this paper, we apply a data-driven approach to the development of privacy-setting recommendations for fitness devices. We first present a fitness data privacy model that we defined to represent users’ privacy preferences in a way that is unambiguous, compliant with the European Union’s General Data Protection Regulation (GDPR), and able to represent both the user and the third party preferences. Our crowdsourced dataset is collected using current scenarios in the fitness domain and used to identify privacy profiles by applying machine learning techniques. We then examine different personal tracking data and user traits which can potentially drive the recommendation of privacy profiles to the users. Finally, a set of privacy-setting recommendation strategies with different guidance styles are designed based on the resulting profiles. Interestingly, our results show several semantic relationships among users’ traits, characteristics, and attitudes that are useful in providing privacy recommendations. Even though several works exist on privacy preference modeling, this paper makes a contribution in modeling privacy preferences for data sharing and processing in the IoT and fitness domain, with specific attention to GDPR compliance. Moreover, the identification of well-identified clusters of preferences and predictors of such clusters is a relevant contribution for user profiling and for the design of interactive recommendation strategies that aim to balance users’ control over their privacy permissions and the simplicity of setting these permissions. 

Research has shown that privacy decisions are affected by heuristic influences such as default settings and framing, and such effects are likely also present in smarthome privacy de- cisions. In this paper we pose the challenge question: How exactly do defaults and framing influence smarthome users’ privacy decisions? We conduct a large-scale scenario-based study with a mixed fractional factorial design, and use sta- tistical analysis and machine learning to investigate these effects. We discuss the implications of our findings for the designers of smarthome privacy-setting interfaces.