Remote Collaborative Knowledge Discovery for Better Understanding of Self-tracking Data

Abstract

Wearable self-tracking devices are an increasingly popular way for people to collect information relevant to their own health and well-being, but maximising the benefits derived from such information is hindered by the complexity of analysing it. To gain deeper insights into their own information generated by such products, a user with no data analysis expertise could collaborate with someone who does have the required knowledge and skills. To achieve such a successful collaboration, several tasks need to be completed: finding a collaborator, negotiating the terms of the collaboration, obtaining the necessary resources, analysing the data and evaluating the results of the analysis. To support the execution of these tasks, we have developed and deployed an online software platform that enables data collectors and owners to find experts and collaborate with them so they can extract additional knowledge from the self-tracking data. The functionality and user interface of the platform are demonstrated by presenting an application scenario where a data owner shares their sleep data with an expert who applies periodicity analysis to discover cyclical patterns from the data.

Keywords

• knowledge discovery
• data mining
• collaborative systems
• personal data
• wearable devices