Improving trust in personal sensing

Abstract

Personal sensing technologies allow subjects to collect and share their personal data with recipients to obtain some benefits. For example, mobile health and wellness (mHealth) devices and applications allow subjects to monitor their own health and lifestyle and share the health and wellness data with their family, friends and peers for emotional support, clinical providers for diagnosis and health coaches for medical advice. Subjects may earn monetary benefits by sharing data with insurance companies, or contribute to the greater good by sharing with researchers. In such scenarios, recipients might be concerned about the accuracy of the data collected, affecting their trust in the data. Malicious subjects and devices can provide fake data. Malfunctioning devices can produce inaccurate data. Careless subjects or delegates (in mHealth scenarios, delegates could be health workers and caretakers who might use devices to collect subject’s data for the recipient) might apply sensors incorrectly and collect inaccurate data. Data could also be collected from wrong subjects or devices deliberately or accidentally. Hence, recipients will need assurances about accuracy of the data and trustworthy behavior of the subjects and delegates who collected the data and the devices that were used in data collection. Recipients might also need metadata to better interpret the meaning of the data. On the other hand, subjects might be concerned about how recipients might use the shared data, affecting their trust in recipients. Subjects might be unsure of what data to share to receive benefits and share more than necessary (and disclose sensitive data unintentionally to unintended recipients). Subjects might be concerned about how their data might be used by the recipients and share less than necessary and not receive benefits from the recipients. Hence, subjects will need assurance about the disclosure of their information, and will want to be aware of how recipients will use their data and what benefits subjects will receive for sharing data with the recipients. The goal of this work is to improve the trust that subjects and recipients have in personal sensing technologies by ensuring that they have control over and can understand how information is collected, shared and used. The paper approaches the problem of improving trust by proposing techniques for increasing awareness and control. We focus on mHealth scenarios, but our solutions are applicable to other personal sensing technologies as well.