Employing Environmental Data and Machine Learning to Improve Mobile Health Receptivity

Theilig, Max-Marcel; Korbel, Jakob J.; Mayer, Gwendolyn; Hoffmann, Christian; Zarnekow, Rüdiger

doi:10.1109/access.2019.2958474

Cited by 6 publications

(5 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An initial step may be to identify the behaviour change techniques or processes of change (Michie et al , 2013; Prochaska et al , 2008) that are necessary to facilitate transition at each stage of the behaviour change journey, and how they may be best incorporated into an app. Additionally, the use of machine learning and environmental data may open up new avenues for more dynamic apps that are better tailored to the user and their individual context (Theilig et al , 2019). While research in this area is currently limited, it offers exciting opportunities to examine app design, user engagement and behaviour change.…”

Section: Discussionmentioning

confidence: 99%

Exploring engagement with health apps: the emerging importance of situational involvement and individual characteristics

Flaherty

McCarthy

Collins

et al. 2021

EJM

View full text Add to dashboard Cite

Purpose Health apps offer a potential approach to support healthier food behaviours but a lack of sufficient engagement may limit effectiveness. This study aims to use a user engagement theoretical lens to examine the factors that influence app engagement over time and may prompt disengagement. Design/methodology/approach A phenomenological exploration of the lived experience was used. Women from a lower socioeconomic background (based on the occupation and employment status of the household’s primary income earner) were randomly assigned to use one of two apps for a minimum of eight weeks. Multiple data collection methods, including accompanied shops, researcher observations, interviews, participant reflective accounts and questionnaires, were used at different time-points to examine engagement. Theoretical thematic analysis was conducted to explore the engagement experience and relevant social, personal and environmental influences. Findings Healthy food involvement appears to drive app engagement. Changes in situational involvement may contribute to fluctuation in engagement intensity over time as the saliency of personal goals change. Negatively valenced engagement dimensions may contribute to the overall expression of engagement. A lack of congruency with personal goals or an imbalance between perceived personal investment and value was expressed as the primary reasons for disengagement. Research limitations/implications Situational involvement may act as a trigger of different engagement phases. There is a need to better distinguish between enduring and situational involvement in engagement research. Practical implications Individual characteristics may shape engagement and propensity for disengagement, which highlights the practical importance of incorporating tailored features into app design. Originality/value Findings broaden the current conceptualisation of engagement within the digital space and prompt a reconsideration of the role of situational involvement and negatively valenced dimensions throughout the engagement process.

show abstract

Section: Discussionmentioning

confidence: 99%

Exploring engagement with health apps: the emerging importance of situational involvement and individual characteristics

Flaherty

McCarthy

Collins

et al. 2021

EJM

View full text Add to dashboard Cite

show abstract

“…Data were collected from 24 activities from six participants who used smartphone devices that had the Fonlog application installed. To determine the initial activity, nineteen activities were taken from the activity list to improve mobile health receptivity by employing environmental data and machine learning [ 84 ], then we distributed the initial activity list to the participants and asked them about their daily activities that were not included in the initial activity, and then we received five additional activities from the participants. We show the activity information in Table 4 .…”

Section: Experimental Evaluationmentioning

confidence: 99%

Optimizing Forecasted Activity Notifications with Reinforcement Learning

Fikry

Inoue

2023

Sensors

View full text Add to dashboard Cite

In this paper, we propose the notification optimization method by providing multiple alternative times as a reminder for a forecasted activity with and without probabilistic considerations for the activity that needs to be completed and needs notification. It is important to consider various factors when sending notifications to people after obtaining the results of the forecasted activity. We should not send notifications only when we have forecasted results because future daily activities are unpredictable. Therefore, it is important to strike a balance between providing useful reminders and avoiding excessive interruptions, especially for low probabilities of forecasted activity. Our study investigates the impact of the low probability of forecasted activity and optimizes the notification time with reinforcement learning. We also show the gaps between forecasted activities that are useful for self-improvement by people for the balance of important tasks, such as tasks completed as planned and additional tasks to be completed. For evaluation, we utilize two datasets: the existing dataset and data we collected in the field with the technology we have developed. In the data collection, we have 23 activities from six participants. To evaluate the effectiveness of these approaches, we assess the percentage of positive responses, user response rate, and response duration as performance criteria. Our proposed method provides a more effective way to optimize notifications. By incorporating the probability level of activity that needs to be done and needs notification into the state, we achieve a better response rate than the baseline, with the advantage of reaching 27.15%, as well as than the other criteria, which are also improved by using probability.

show abstract

“…The resulting dataset contains a total of 300 minutes of measurements, where each data sample is composed of 228 features extracted from both physical and virtual sensors, including accelerometer, gyroscope, user location, and phone state (e.g., app status, battery state, and Wi-Fi availability). Thanks to the variety of the collected data, ExtraSensory has been used in several research works for the evaluation of hu-man activity recognition systems [30,31], personalization of machine learning models [32], and behavioral assessments for e-health applications [33].…”

Section: Related Workmentioning

confidence: 99%

MyDigitalFootprint: An extensive context dataset for pervasive computing applications at the edge

Campana¹,

Delmastro²

2021

Pervasive and Mobile Computing

View full text Add to dashboard Cite

Employing Environmental Data and Machine Learning to Improve Mobile Health Receptivity

Cited by 6 publications

References 60 publications

Exploring engagement with health apps: the emerging importance of situational involvement and individual characteristics

Exploring engagement with health apps: the emerging importance of situational involvement and individual characteristics

Optimizing Forecasted Activity Notifications with Reinforcement Learning

MyDigitalFootprint: An extensive context dataset for pervasive computing applications at the edge

Contact Info

Product

Resources

About