Living a sedentary lifestyle is one of the major causes of numerous health problems. To encourage employees to lead a less sedentary life, the Hanze University started a health promotion program. One of the interventions in the program was the use of an activity tracker to record participants' daily step count. The daily step count served as input for a fortnightly coaching session. In this paper, we investigate the possibility of automating part of the coaching procedure on physical activity by providing personalized feedback throughout the day on a participant's progress in achieving a personal step goal. The gathered step count data was used to train eight different machine learning algorithms to make hourly estimations of the probability of achieving a personalized, daily steps threshold. In 80% of the individual cases, the Random Forest algorithm was the best performing algorithm (mean accuracy = 0.93, range = 0.88–0.99, and mean F1-score = 0.90, range = 0.87–0.94). To demonstrate the practical usefulness of these models, we developed a proof-of-concept Web application that provides personalized feedback about whether a participant is expected to reach his or her daily threshold. We argue that the use of machine learning could become an invaluable asset in the process of automated personalized coaching. The individualized algorithms allow for predicting physical activity during the day and provides the possibility to intervene in time.

One of the promises of the experience sampling methodology (ESM) is that it could be used to identify relevant targets for treatment, based on a statistical analysis of an individual’s emotions, cognitions and behaviors in everyday-life. A requisite for clinical implementation is that outcomes of person-centered analyses are not wholly contingent on the researcher performing them. To evaluate how much researchers vary in their analytical approach and to what degree outcomes vary based on analytical choices, we crowdsourced the analysis of one individual patient’s ESM data to 12 prominent research teams, asking them what symptom(s) they would advise the treating clinician to target in subsequent treatment. The dataset was from a 25-year-old male with a primary diagnosis of major depressive disorder and comorbid generalized anxiety disorder, who completed momentary assessments related to depression and anxiety psychopathology prior to psychotherapy. Variation was evident at different stages of the analysis, from preprocessing steps (e.g., variable selection, clustering, handling of missing data) to the type of statistics. Most teams did include a type of vector autoregressive model, which examines relations between variables (e.g., symptoms) over time. Although most teams were confident their selected targets would provide useful information to the clinician, not one advice was similar: both the number (0-16) and nature of selected targets varied widely. This study makes transparent that the selection of treatment targets based on personalized models using ESM data is currently highly conditional on subjective analytical choices and highlights key methodological issues that need to be addressed in moving toward clinical implementation. Research proposal, data and materials: osf.io/h3djy/

We present the u-can-act platform, a tool that we developed to study the individual processes of early school leaving and the preventative actions that mentors take to steer these processes in the right direction. Early school leaving is a significant problem, particularly in vocational education, and can have severe consequences for both the individual and society. However, the prevention of early school leaving is hampered by a mismatch between research and practice: research tends to focus on identifying risk factors using group averages and cross-sectional studies, while practitioners focus on intervening in individual processes. We aim to help solve this mismatch with our project u-can-act. In this project we have developed a platform that helps to gain insight into both the individual processes that precede early school leaving as well as the actions that mentors take to prevent it. In this paper we introduce the u-can-act platform, which consists of three technology-based, reusable methodological innovations. Specifically, our innovations concern: (i) an open source web application for longitudinal personalized data-collection, (ii) an automated study protocol that optimizes adherence in a difficult target group (adolescents at risk for early school leaving), and (iii) a technologically assisted coupling between mentor and student that allows us to study dyadic interactions over time. We present performance results of our platform, including participant adherence, the behavior of the questionnaire items over time, and the way that our web application is experienced by the participants. We conclude that our innovative platform is successful in collecting multi-informant time-series data on intervention processes among students in vocational education, both for at-risk students and control students, and for their mentors. Moreover, our platform is suitable for broader applications: it can be used to study any malleable individual process including the efforts of a second individual who aims to influence this process. Because of the unique insights that the u-can-act platform is able to generate, the platform may ultimately contribute to solving the mismatch between research and practice, and to more effective interventions in individual processes.

Precise fine-grained office occupancy detection can be exploited for energy savings in buildings. Based on such information one can optimally regulate lighting and climatization based on the actual presence and absence of users. Conventional approaches are based on movement detection, which are cheap and easy to deploy, but are imprecise and offer coarse information. We propose a power monitoring system as a source of occupancy information.The approach is based on sub-metering at the level of room circuit breakers. The proposed method tackles the problem of indoor, office occupancy detection based on statistical approaches based on power consumption information thus contributing to building context awareness which, in turn, is a crucial stepping stone for energy efficient buildings. The key advantage of the proposed approach is to be low intrusive, especially when compared with image or tag based solutions, while still being sufficiently precise in its classification. Such classification is based on nearest neighbors and neural networks machine learning approaches, both in sequential and non-sequential implementations. To test the viability, precision, and saving potential of the proposed approach we deploy in an actual office over several months. We find that the room-level sub- metering can acquire precise, fine-grained occupancy context for up to three people, with averaged kappa measures of 93-95% using either the nearest neighbors or neural networks based approaches.