About the book
Self tracking has become part of a modern lifestyle; wearables and smartphones support self tracking in an easy fashion and change our behavior such as in the health sphere. The amount of data generated by these devices is so overwhelming that it is difficult to get useful insight from it. Luckily, in the domain of artificial intelligence techniques exist that can help out here: machine learning approaches are well suited to assist a
nd enable one to analyze this type of data. While there are ample books that explain machine learning techniques, self-tracking data comes with its own difficulties that require dedicated techniques such as learning over time and across users. In this book, we will explain the complete loop to effectively useself-tracking data for machine learning; from cleaning the data, the identification of features, finding clusters in the data, algorithms to create predictions of values for the present and future, and learning how to provide feedback to users based on their tracking data. All concepts we explain draw from state-of-the-art scientific literature. To illustrate all approaches, we use a case study of a rich self-tracking dataset obtained from the crowdsignals platform. While the book is focused on the self-tracking data, the techniques explained are more widely applicable to sensory data in general, making it useful for a wider audience.
About the authors
Mark Hoogendoorn is an Assistant Professor in the Computational Intelligence Group at the Department of Computer Science of the Vrije Universiteit Amsterdam. He obtained his PhD degree in Artificial Intelligence at the same university in 2007. After completing his PhD he has been a Postdoctoral Researcher at the University of Minnesota, Department of Computer Science and Engineering after which he started as an Assistant Professor at the VU. In 2015 he was a Visiting Scientist at the Computer Science and Artificial Intelligence lab (CSAIL) of the Massachusetts Institute for Technology. In his research, he mainly focuses on predictive modeling and personalization using AI techniques, applied to various domains such as health care. He has published over 100 peer-reviewed research papers in the area of AI and has obtained funding for his research from various national and international funding agencies, including EU FP7 and H2020 projects and national funding (STW) as well as funding from industrial partners. For the EU FP7 ICT4Depression project he was the joint coordinator. Dr. Hoogendoorn has been involved in the organization of several prominent AI related conferences and served on the program committees of a variety of conferences as well. He is currently on the board of directors of ISRII, an international organization for researchers involved in the development of innovative interventions for mental health.
Burkhardt Funk is a full professor for information systems at Leuphana University Lüneburg/ Germany. His research interests encompass building statistical models and decision support systems, based on methods from machine learning and data mining, in a variety of application domains such as e-Commerce, ad-tech, and e-Mental-Health. He has co-authored more than 70 scientific articles and acquired project funding of more than 2 Mio. EUR (personal share). As a Co-PI of two large scale, EU funded research projects on e-Mental-Health he contributes to the understanding of mental disorders using predictive modeling techniques. Dr. Funk has been a visiting scientist at the University of Virginia and Stanford University in 2016. With respect to teaching he is proud to be one of the driving forces behind a newly established, international Master program in Data Science at Leuphana University.
Table of contents
For the table of contents, please click here.
Buying the book
For ordering the book, please go to the Springer shop or Amazon. Some (university) libraries have subscriptions that provide full online access to the book via Springer.