Insufficient physical activity presents a significant hazard to overall health, with sedentary lifestyles linked to a variety of health issues. Monitoring physical activity levels allows the recognition of patterns of sedentary behavior and the provision of coaching to meet the recommended physical activity standards. In this paper, we aim to address the problem of reducing the time consuming process of fitting classifiers when generating personalized models for a coaching application. The proposed approach consists of evaluating the effects of clustering participants based on their walking patterns and then recommending a unique model for each group. Each model consists of a random forest classifier with a different number of estimators each. The resulting approach reduces the fitting time considerably while keeping nearly the same classification performance as personalized models.
DOCUMENT
Objectives: Emotional eating is recognized as a potential contributor to weight gain. Emotional eaters often hide their problems because of feelings of shame about their behavior, making it challenging to provide them with the necessary support. The introduction of a virtual coach might offer a potential solution in assisting them. To find out whether emotional eaters are receptive to online personalized coaching, we presented emotional eaters with two essential proto-typical problem situations for emotional eaters: “experiencing cravings” and “after giving in to cravings,” and asked them whether they preferred one of the three coaching strategies presented: Validating, Focus-on-Change and Dialectical.Methods: An experimental vignette study (2 × 3 design) was carried out. The vignettes featured two distinct personas, each representing one of the two common problem scenarios experienced by emotional eaters, along with three distinct coaching strategies for each scenario. To identify potential predictors for recognition of problem situations, questionnaires on emotional eating (DEBQ), personality traits (Big-5), well-being (PANAS), and BMI were administrated.Results: A total of 62% of the respondents identified themselves with “after giving in to cravings” and 47% with “experiencing cravings.” BMI, emotional eating and emotional stability appeared to be predictors in recognizing both the problem situations. In “experiencing cravings,” the participating women preferred Dialectical and the Validation coaching strategies. In the “after giving in to cravings” condition, they revealed a preference for the Dialectical and the Focus-on-Change coaching strategies.Conclusion: Using vignettes allowed a less threatening way of bringing up sensitive topics for emotional eaters. The personas representing the problem situations were reasonably well recognized. To further enhance this recognition, it is important for the design and content of the personas to be even more closely related to the typical problem scenarios of emotional eaters, rather than focusing on physical characteristics or social backgrounds. This way, users may be less distracted by these factors. With the knowledge gained about the predictors that may influence recognition of the problem situations, design for coaching can be more customized. The participants represented individuals with high emotional eating levels, enhancing external validity.
DOCUMENT
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.
DOCUMENT
