Many people are concerned about how their personal data is used for online behavioral advertising (OBA). Ad targeting explanations have been proposed as a way to reduce this concern by improving transparency. However, it is unclear when and why people might want ad targeting explanations. Without this insight, we run the risk of designing explanations that do not address real concerns. To bridge this gap, we conducted a four-week, mixed-methods field study with 60 participants to understand when and why people want targeting explanations for the ads they actually encountered while browsing the web. We found that users wanted explanations for around 30% of the 4,251 ads we asked them about during the study, and that subjective perceptions of how their personal data was collected and shared were highly correlated with when users wanted ad explanations. Often, users wanted these explanations to confirm or deny their own preconceptions about how their data was collected or the motives of advertisers. A key upshot of our work is that one-size-fits-all approaches to ad explanations are likely to fail at addressing people’s lived concerns about ad targeting; instead, more personalized explanations are needed.

Trust is a fundamental prerequisite in the growth and sustainability of sharing economy platforms. Many of such platforms rely on actions that require trust to take place, such as entering a stranger’s car or sleeping at a stranger’s place. For this reason, understanding, measuring, and tracking trust can be of great benefit to such platforms, enabling them to identify trust behaviors, both online and offline, and identify groups which may benefit from trust-building interventions. In this work, we present the design and evaluation of a behavioral framework to measure a user’s propensity to trust others on Airbnb. We conducted an online experiment with 4,499 Airbnb users in the form of an investment game in order to capture users’ propensity to trust other users on Airbnb. Then, we used the experimental data to generate both explanatory and predictive models of trust propensity. Our contribution is a framework that can be used to measure trust propensity in sharing economy platforms via online and offline signals. We discuss which affordances need to be in place so that sharing economy platforms can get signals of trust, in addition to how such a framework can be used to inform design around trust in the short and long term.

The increased use of machine learning in recent years led to large volumes of data being manually labeled via crowdsourcing microtasks completed by humans. This brought about dehumanization effects, namely, when task requesters overlook the humans behind the task, leading to issues of ethics (e.g., unfair payment) and amplification of human biases, which are transferred into training data and affect machine learning in the real world. We propose a framework that allocates microtasks considering human factors of workers such as demographics and compensation. We deployed our framework to a popular crowdsourcing platform and conducted experiments with 1,919 workers collecting 160,345 human judgments. By routing microtasks to workers based on demographics and appropriate pay, our framework mitigates biases in the contributor sample and increases the hourly pay given to contributors. We discuss potential extensions and how it can promote transparency in crowdsourcing.

Abstract Smart home devices challenge a long-held notion that the home is a private and protected place. With this in mind, many developers market their products with a focus on privacy in order to gain user trust, yet privacy tensions arise with the growing adoption of these devices and the risk of inappropriate data practices in the smart home (e.g., secondary use of collected data). Therefore, it is important for developers to consider individual user preferences and how they would change under varying circumstances, in order to identify actionable steps towards developing user trust and exercising privacy-preserving data practices. To help achieve this, we present the design and evaluation of machine learning models that predict (1) personalized allow/deny decisions for different information flows involving various attributes, purposes, and devices (AUC .868), (2) what circumstances may change original decisions (AUC .899), and (3) how much (US dollars) one may be willing to pay or receive in exchange for smart home privacy (RMSE 12.459). We show how developers can use our models to derive actionable steps toward privacy-preserving data practices in the smart home.

Visually impaired users face various challenges in web authentication. We designed UniPass, an accessible password manager for visually impaired users based on a smart device. To evaluate UniPass, we tested and compared UniPass with two commercial password managers: LastPass, a popular password manager and StrongPass, a smart device-based password manager. Our study results of ten users, six blind and four with low vision, suggest that password managers are a promising authentication approach for visually impaired users. Participants using UniPass had the highest task completion rate and took the shortest time to complete an authentication related task. Furthermore, the majority (seven out of ten) of our participants preferred UniPass over LastPass and StrongPass.

This paper briefly describes a proposed interaction workflow that is currently being developed as part of a research effort towards providing better solutions for accessible authentication, strongly guided by contextual inquiry and evidence-based guidelines. The approach described herein is being developed and tested to be foundations for tests and findings of the research, consequently evolving along the research progress towards providing a scalable, deployable, secure, usable for everyone, and last, but not least, privacy preserving platform for web authentication.