Suparna Ghanvatkar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Suparna Ghanvatkar has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Vaibhav Rajan
Department of Information Systems and Analytics, National University of Singapore, Singapore
6 shared publications - Atreyi Kankanhalli
Department of Information Systems and Analytics, School of Computing National University of Singapore, Singapore
4 shared publications - Saurabh Chaudhari
Department of Information Systems and Analytics School of Computing National University of Singapore Singapore Singapore
2 shared publications Belinda Bugay Flores
Autism Recovery Network
1 shared publicationCharmaine Joan P Mendoza
Autism Recovery Network
1 shared publicationKonstantinos Trakakis
Spectrum Learning Analytics
1 shared publication
Latest Publications
- 23 July 2026
- Published by Oxford University Press (OUP)
From Tacit Knowledge to Data Representations for Information Systems in Autism Therapy Planning
Autism therapy planning, constrained by heterogeneity and resource limitations, is a judgment-intensive process requiring continual decisions about what to teach, how to sequence skills, and how to adapt instruction to individual learner needs. Contemporary information systems capture assessment and progress data but typically treat recorded scores as neutral indicators of learner change. Drawing on routinely collected longitudinal assessment records together with qualitative observation of therapy planning practices in an Applied Behavior Analysis center, this article shows that recorded assessment trajectories reflect not only learner development but also tacit and partially specified planning judgments that shape how data are generated. Quantitative analyses of the longitudinal records reveal consistent yet practice-sensitive patterns, and qualitative coding makes these planning judgments explicit via an entity-relationship schema and a five-stage process model. While grounded in a single setting, this work contributes a conceptual and empirical foundation for more cautious and context-aware reuse of autism therapy data in data-driven systems, including AI-enabled information systems, where recorded observations may be shaped by professional judgment.
- 23 July 2026
- Published by Oxford University Press (OUP)
From Tacit Knowledge to Data Representations for Information Systems in Autism Therapy Planning
Autism therapy planning, constrained by heterogeneity and resource limitations, is a judgment-intensive process requiring continual decisions about what to teach, how to sequence skills, and how to adapt instruction to individual learner needs. Contemporary information systems capture assessment and progress data but typically treat recorded scores as neutral indicators of learner change. Drawing on routinely collected longitudinal assessment records together with qualitative observation of therapy planning practices in an Applied Behavior Analysis center, this article shows that recorded assessment trajectories reflect not only learner development but also tacit and partially specified planning judgments that shape how data are generated. Quantitative analyses of the longitudinal records reveal consistent yet practice-sensitive patterns, and qualitative coding makes these planning judgments explicit via an entity-relationship schema and a five-stage process model. While grounded in a single setting, this work contributes a conceptual and empirical foundation for more cautious and context-aware reuse of autism therapy data in data-driven systems, including AI-enabled information systems, where recorded observations may be shaped by professional judgment.
- 25 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 28 (7) , 4269-4280
Evaluating Explanations From AI Algorithms for Clinical Decision-Making: A Social Science-Based Approach
Explainable Artificial Intelligence (XAI) techniques generate explanations for predictions from AI models. These explanations can be evaluated for (i) faithfulness to the prediction, i.e., its correctness about the reasons for prediction, and (ii) usefulness to the user. While there are metrics to evaluate faithfulness, to our knowledge, there are no automated metrics to evaluate the usefulness of explanations in the clinical context. Our objective is to develop a new metric to evaluate usefulness of AI explanations to clinicians. Usefulness evaluation needs to consider both (a) how humans generally process explanations and (b) clinicians' specific requirements from explanations presented by clinical decision support systems (CDSS). Our new scoring method can evaluate the usefulness of explanations generated by any XAI method that provides importance values for the input features of the prediction model. Our method draws on theories from social science to gauge usefulness, and uses literature-derived biomedical knowledge graphs to quantify support for the explanations from clinical literature. We evaluate our method in a case study on predicting onset of sepsis in intensive care units. Our analysis shows that the scores obtained using our method corroborate with independent evidence from clinical literature and have the required qualities expected from such a metric. Thus, our method can be used to evaluate and select useful explanations from a diverse set of XAI techniques in clinical contexts, making it a fundamental tool for future research in the design of AI-driven CDSS.
- 25 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 28 (7) , 4269-4280
Evaluating Explanations From AI Algorithms for Clinical Decision-Making: A Social Science-Based Approach
Explainable Artificial Intelligence (XAI) techniques generate explanations for predictions from AI models. These explanations can be evaluated for (i) faithfulness to the prediction, i.e., its correctness about the reasons for prediction, and (ii) usefulness to the user. While there are metrics to evaluate faithfulness, to our knowledge, there are no automated metrics to evaluate the usefulness of explanations in the clinical context. Our objective is to develop a new metric to evaluate usefulness of AI explanations to clinicians. Usefulness evaluation needs to consider both (a) how humans generally process explanations and (b) clinicians' specific requirements from explanations presented by clinical decision support systems (CDSS). Our new scoring method can evaluate the usefulness of explanations generated by any XAI method that provides importance values for the input features of the prediction model. Our method draws on theories from social science to gauge usefulness, and uses literature-derived biomedical knowledge graphs to quantify support for the explanations from clinical literature. We evaluate our method in a case study on predicting onset of sepsis in intensive care units. Our analysis shows that the scores obtained using our method corroborate with independent evidence from clinical literature and have the required qualities expected from such a metric. Thus, our method can be used to evaluate and select useful explanations from a diverse set of XAI techniques in clinical contexts, making it a fundamental tool for future research in the design of AI-driven CDSS.
- Open Access
Evaluating Explanations from AI Algorithms for Clinical Decision-Making: A Social Science-based Approach
Explainable Artificial Intelligence (XAI) techniques generate explanations for predictions from AI models. These explanations can be evaluated for (i) faithfulness to the prediction, i.e., its correctness about the reasons for prediction, and (ii) usefulness to the user. While there are metrics to evaluate faithfulness, to our knowledge, there are no automated metrics to evaluate the usefulness of explanations in the clinical context. Our objective is to develop a new metric to evaluate usefulness of AI explanations to clinicians. Usefulness evaluation needs to consider both (a) how humans generally process explanations and (b) clinicians’ specific requirements from explanations presented by clinical decision support systems (CDSS). Our new scoring method can evaluate the usefulness of explanations generated by any XAI method that provides importance values for the input features of the prediction model. Our method draws on theories from social science to gauge usefulness, and uses literature-derived biomedical knowledge graphs to quantify support for the explanations from clinical literature. We evaluate our method in a case study on predicting onset of sepsis in intensive care units. Our analysis shows that the scores obtained using our method corroborate with independent evidence from clinical literature and have the required qualities expected from such a metric. Thus, our method can be used to evaluate and select useful explanations from a diverse set of XAI techniques in clinical contexts, making it a fundamental tool for future research in the design of AI-driven CDSS.
Open AccessEvaluating Explanations from AI Algorithms for Clinical Decision-Making: A Social Science-based Approach
Explainable Artificial Intelligence (XAI) techniques generate explanations for predictions from AI models. These explanations can be evaluated for (i) faithfulness to the prediction, i.e., its correctness about the reasons for prediction, and (ii) usefulness to the user. While there are metrics to evaluate faithfulness, to our knowledge, there are no automated metrics to evaluate the usefulness of explanations in the clinical context. Our objective is to develop a new metric to evaluate usefulness of AI explanations to clinicians. Usefulness evaluation needs to consider both (a) how humans generally process explanations and (b) clinicians’ specific requirements from explanations presented by clinical decision support systems (CDSS). Our new scoring method can evaluate the usefulness of explanations generated by any XAI method that provides importance values for the input features of the prediction model. Our method draws on theories from social science to gauge usefulness, and uses literature-derived biomedical knowledge graphs to quantify support for the explanations from clinical literature. We evaluate our method in a case study on predicting onset of sepsis in intensive care units. Our analysis shows that the scores obtained using our method corroborate with independent evidence from clinical literature and have the required qualities expected from such a metric. Thus, our method can be used to evaluate and select useful explanations from a diverse set of XAI techniques in clinical contexts, making it a fundamental tool for future research in the design of AI-driven CDSS.
- Open Access
Graph-Based Patient Representation for Multimodal Clinical Data: Addressing Data Heterogeneity
Patient data comprises of different modalities like clinical notes, lab results, and radiological investigations. Predictive modeling on this patient data is challenging due to the heterogeneity among patients and modalities captured, such as no ECG recordings for a few patients. Ensuring model applicability to all admitted patients requires addressing three key factors: i) handling modalities at disparate time-scales (e.g., once in a few hours medication and 125 Hz ECG need to be handled by same model), ii) handling missing modality (e.g., ECG not recorded), and iii) modeling temporal interactions between modalities over time (e.g., the ECG fluctuation triggered some lab test order). Existing literature often doesn’t simultaneously address these requirements. Therefore, we propose a novel patient representation approach inspired by clinical workflows, representing each patient as a graph. We categorize patient data into two main components: observations and care team actions, allowing cross-modality temporal interaction when actions consider the previous observations. We define observation nodes to capture modality-specific data within the time between two actions, and action nodes capture data of the actions undertaken, with edges to capture the temporal dependencies. To address missing modalities and time-scale disparities, we define node types for different modalities and use modality-specific representation for the nodes; implying a missing modality is equivalent to a missing node type in the graph. Aligned with clinical workflows, this patient-graph representation aims to enhance the practicality of predictive systems for various healthcare tasks, from mortality risk assessment to medication recommendations, thereby improving clinical decision support.
Open AccessGraph-Based Patient Representation for Multimodal Clinical Data: Addressing Data Heterogeneity
Patient data comprises of different modalities like clinical notes, lab results, and radiological investigations. Predictive modeling on this patient data is challenging due to the heterogeneity among patients and modalities captured, such as no ECG recordings for a few patients. Ensuring model applicability to all admitted patients requires addressing three key factors: i) handling modalities at disparate time-scales (e.g., once in a few hours medication and 125 Hz ECG need to be handled by same model), ii) handling missing modality (e.g., ECG not recorded), and iii) modeling temporal interactions between modalities over time (e.g., the ECG fluctuation triggered some lab test order). Existing literature often doesn’t simultaneously address these requirements. Therefore, we propose a novel patient representation approach inspired by clinical workflows, representing each patient as a graph. We categorize patient data into two main components: observations and care team actions, allowing cross-modality temporal interaction when actions consider the previous observations. We define observation nodes to capture modality-specific data within the time between two actions, and action nodes capture data of the actions undertaken, with edges to capture the temporal dependencies. To address missing modalities and time-scale disparities, we define node types for different modalities and use modality-specific representation for the nodes; implying a missing modality is equivalent to a missing node type in the graph. Aligned with clinical workflows, this patient-graph representation aims to enhance the practicality of predictive systems for various healthcare tasks, from mortality risk assessment to medication recommendations, thereby improving clinical decision support.
- Open Access
- 28 February 2022
- Published by JMIR Publications Inc. in JMIR mHealth and uHealth
- Vol. 10 (2) , e31327
Personalization of Intervention Timing for Physical Activity: Scoping Review
Background: The use of sensors in smartphones, smartwatches, and wearable devices has facilitated the personalization of interventions to increase users’ physical activity (PA). Recent research has focused on evaluating the effects of personalized interventions in improving PA among users. However, it is critical to deliver the intervention at an appropriate time to each user to increase the likelihood of adoption of the intervention. Earlier review studies have not focused on the personalization of intervention timing for increasing PA. Objective: This review aims to examine studies of information technology–based PA interventions with personalized intervention timing (PIT); identify inputs (eg, user location) used by the system for generating the PIT, the techniques and methods used for generating the PIT, the content of the PA intervention, and delivery mode of the intervention; and identify gaps in existing literature and suggest future research directions. Methods: A scoping review was undertaken using PsycINFO, PubMed, Scopus, and Web of Science databases based on a structured search query. The main inclusion criteria were as follows: the study aimed to promote PA, included some form of PIT, and used some form of information technology for delivery of the intervention to the user. If deemed relevant, articles were included in this review after removing duplicates and examining the title, abstract, and full text of the shortlisted articles. Results: The literature search resulted in 18 eligible studies. In this review, 72% (13/18) of the studies focused on increasing PA as the primary objective, whereas it was the secondary focus in the remaining studies. The inputs used to generate the PIT were categorized as user preference, activity level, schedule, location, and predicted patterns. On the basis of the intervention technique, studies were classified as manual, semiautomated, or automated. Of these, the automated interventions were either knowledge based (based on rules or guidelines) or data driven. Of the 18 studies, only 6 (33%) evaluated the effectiveness of the intervention and reported positive outcomes. Conclusions: This work reviewed studies on PIT for PA interventions and identified several aspects of the interventions, that is, inputs, techniques, contents, and delivery mode. The reviewed studies evaluated PIT in conjunction with other personalization approaches such as activity recommendation, with no study evaluating the effectiveness of PIT alone. On the basis of the findings, several important directions for future research are also highlighted in this review.
- 28 February 2022
- Published by JMIR Publications Inc. in JMIR mHealth and uHealth
- Vol. 10 (2) , e31327
Open AccessPersonalization of Intervention Timing for Physical Activity: Scoping Review
Background: The use of sensors in smartphones, smartwatches, and wearable devices has facilitated the personalization of interventions to increase users’ physical activity (PA). Recent research has focused on evaluating the effects of personalized interventions in improving PA among users. However, it is critical to deliver the intervention at an appropriate time to each user to increase the likelihood of adoption of the intervention. Earlier review studies have not focused on the personalization of intervention timing for increasing PA. Objective: This review aims to examine studies of information technology–based PA interventions with personalized intervention timing (PIT); identify inputs (eg, user location) used by the system for generating the PIT, the techniques and methods used for generating the PIT, the content of the PA intervention, and delivery mode of the intervention; and identify gaps in existing literature and suggest future research directions. Methods: A scoping review was undertaken using PsycINFO, PubMed, Scopus, and Web of Science databases based on a structured search query. The main inclusion criteria were as follows: the study aimed to promote PA, included some form of PIT, and used some form of information technology for delivery of the intervention to the user. If deemed relevant, articles were included in this review after removing duplicates and examining the title, abstract, and full text of the shortlisted articles. Results: The literature search resulted in 18 eligible studies. In this review, 72% (13/18) of the studies focused on increasing PA as the primary objective, whereas it was the secondary focus in the remaining studies. The inputs used to generate the PIT were categorized as user preference, activity level, schedule, location, and predicted patterns. On the basis of the intervention technique, studies were classified as manual, semiautomated, or automated. Of these, the automated interventions were either knowledge based (based on rules or guidelines) or data driven. Of the 18 studies, only 6 (33%) evaluated the effectiveness of the intervention and reported positive outcomes. Conclusions: This work reviewed studies on PIT for PA interventions and identified several aspects of the interventions, that is, inputs, techniques, contents, and delivery mode. The reviewed studies evaluated PIT in conjunction with other personalization approaches such as activity recommendation, with no study evaluating the effectiveness of PIT alone. On the basis of the findings, several important directions for future research are also highlighted in this review.
- Open Access
- 17 June 2021
- Published by JMIR Publications Inc. in JMIR Preprints
Personalization of Intervention Timing for Physical Activity: Scoping Review (Preprint)
BACKGROUND The use of sensors in smartphones, smartwatches, and wearable devices has facilitated the personalization of interventions to increase users’ physical activity (PA). Recent research has focused on evaluating the effects of personalized interventions in improving PA among users. However, it is critical to deliver the intervention at an appropriate time to each user to increase the likelihood of adoption of the intervention. Earlier review studies have not focused on the personalization of intervention timing for increasing PA. OBJECTIVE This review aims to examine studies of information technology–based PA interventions with personalized intervention timing (PIT); identify inputs (eg, user location) used by the system for generating the PIT, the techniques and methods used for generating the PIT, the content of the PA intervention, and delivery mode of the intervention; and identify gaps in existing literature and suggest future research directions. METHODS A scoping review was undertaken using PsycINFO, PubMed, Scopus, and Web of Science databases based on a structured search query. The main inclusion criteria were as follows: the study aimed to promote PA, included some form of PIT, and used some form of information technology for delivery of the intervention to the user. If deemed relevant, articles were included in this review after removing duplicates and examining the title, abstract, and full text of the shortlisted articles. RESULTS The literature search resulted in 18 eligible studies. In this review, 72% (13/18) of the studies focused on increasing PA as the primary objective, whereas it was the secondary focus in the remaining studies. The inputs used to generate the PIT were categorized as user preference, activity level, schedule, location, and predicted patterns. On the basis of the intervention technique, studies were classified as manual, semiautomated, or automated. Of these, the automated interventions were either knowledge based (based on rules or guidelines) or data driven. Of the 18 studies, only 6 (33%) evaluated the effectiveness of the intervention and reported positive outcomes. CONCLUSIONS This work reviewed studies on PIT for PA interventions and identified several aspects of the interventions, that is, inputs, techniques, contents, and delivery mode. The reviewed studies evaluated PIT in conjunction with other personalization approaches such as activity recommendation, with no study evaluating the effectiveness of PIT alone. On the basis of the findings, several important directions for future research are also highlighted in this review.
- 17 June 2021
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessPersonalization of Intervention Timing for Physical Activity: Scoping Review (Preprint)
BACKGROUND The use of sensors in smartphones, smartwatches, and wearable devices has facilitated the personalization of interventions to increase users’ physical activity (PA). Recent research has focused on evaluating the effects of personalized interventions in improving PA among users. However, it is critical to deliver the intervention at an appropriate time to each user to increase the likelihood of adoption of the intervention. Earlier review studies have not focused on the personalization of intervention timing for increasing PA. OBJECTIVE This review aims to examine studies of information technology–based PA interventions with personalized intervention timing (PIT); identify inputs (eg, user location) used by the system for generating the PIT, the techniques and methods used for generating the PIT, the content of the PA intervention, and delivery mode of the intervention; and identify gaps in existing literature and suggest future research directions. METHODS A scoping review was undertaken using PsycINFO, PubMed, Scopus, and Web of Science databases based on a structured search query. The main inclusion criteria were as follows: the study aimed to promote PA, included some form of PIT, and used some form of information technology for delivery of the intervention to the user. If deemed relevant, articles were included in this review after removing duplicates and examining the title, abstract, and full text of the shortlisted articles. RESULTS The literature search resulted in 18 eligible studies. In this review, 72% (13/18) of the studies focused on increasing PA as the primary objective, whereas it was the secondary focus in the remaining studies. The inputs used to generate the PIT were categorized as user preference, activity level, schedule, location, and predicted patterns. On the basis of the intervention technique, studies were classified as manual, semiautomated, or automated. Of these, the automated interventions were either knowledge based (based on rules or guidelines) or data driven. Of the 18 studies, only 6 (33%) evaluated the effectiveness of the intervention and reported positive outcomes. CONCLUSIONS This work reviewed studies on PIT for PA interventions and identified several aspects of the interventions, that is, inputs, techniques, contents, and delivery mode. The reviewed studies evaluated PIT in conjunction with other personalization approaches such as activity recommendation, with no study evaluating the effectiveness of PIT alone. On the basis of the findings, several important directions for future research are also highlighted in this review.
- Open Access
- 16 January 2019
- Published by JMIR Publications Inc. in JMIR mHealth and uHealth
- Vol. 7 (1) , e11098
User Models for Personalized Physical Activity Interventions: Scoping Review
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: Fitness devices have spurred the development of applications that not only monitor physical activity (PA) but also aim to motivate users, through interventions, to increase their PA. Personalization in the interventions is essential as the target users are diverse with respect to their activity levels, requirements, preferences, and behaviour. Objective: This review aims to: 1) identify different kinds of personalization in interventions for promoting PA among any type of user group; 2) identify user models used for providing personalization; and 3) identify gaps in the current literature and suggest future research directions. Methods: A scoping review was undertaken by searching the databases PsycINFO, PubMed, Scopus and Web of Science databases. The main inclusion criteria were: 1) studies that aimed to promote PA among target users as at least one of their objectives; 2) studies that had user personalization, with the intention of promoting PA (e.g., activity recommendations or motivational messages) through technology-based interventions; 3) studies that described user models for personalization. Results: The literature search resulted in 49 eligible studies. Of these, 67% (33 studies) focused solely on increasing PA, while the remaining studies had other objectives, such as maintaining healthy lifestyle (8 studies), weight loss management (6 studies), and rehabilitation (2 studies) as well. The reviewed studies provide personalization in six categories related to recommendation and feedback: goal recommendation, activity recommendation, fitness partner recommendation, educational content, motivational content, and intervention timing. With respect to the mode of generation, interventions were found to be semi-automated, or automatic. Of these, the automatic interventions were either knowledge-based or data-driven or both. User models in the studies were constructed with parameters from five categories: PA profile, demographics, medical data, behaviour change technique (BCT) parameters, and contextual information. Only 27 of the eligible studies evaluated the interventions for improvement in PA and 16 of these concluded that the interventions to increase PA are more effective when they are personalized. Conclusions: This scoping review investigates personalization of technology-based interventions, in the form of recommendations or feedback for increasing PA. Based on the review and gaps identified, research directions for improving the efficacy of personalized interventions are proposed. First, data-driven prediction techniques can facilitate effective personalization of automated and semi-automated interventions. Second, use of BCTs in automated interventions, and in combination with PA guidelines, are yet to be explored, and preliminary studies in this direction are promising. Third, systems with automated interventions also need to be suitably adapted to serve specific needs of patients with clinical conditions. Fourth, previous user models focus on single metric evaluations of PA instead of a, potentially more effective, holistic and multidimensional view. Fifth, with the widespread adoption of activity monitoring devices and smartphones, personalized and dynamic user models can be created using available user data, including users’ social profile. Finally, the long-term effects of such interventions, as well as the technology medium used for the interventions, need to be evaluated rigorously.
- 16 January 2019
- Published by JMIR Publications Inc. in JMIR mHealth and uHealth
- Vol. 7 (1) , e11098
Open AccessUser Models for Personalized Physical Activity Interventions: Scoping Review
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: Fitness devices have spurred the development of applications that not only monitor physical activity (PA) but also aim to motivate users, through interventions, to increase their PA. Personalization in the interventions is essential as the target users are diverse with respect to their activity levels, requirements, preferences, and behaviour. Objective: This review aims to: 1) identify different kinds of personalization in interventions for promoting PA among any type of user group; 2) identify user models used for providing personalization; and 3) identify gaps in the current literature and suggest future research directions. Methods: A scoping review was undertaken by searching the databases PsycINFO, PubMed, Scopus and Web of Science databases. The main inclusion criteria were: 1) studies that aimed to promote PA among target users as at least one of their objectives; 2) studies that had user personalization, with the intention of promoting PA (e.g., activity recommendations or motivational messages) through technology-based interventions; 3) studies that described user models for personalization. Results: The literature search resulted in 49 eligible studies. Of these, 67% (33 studies) focused solely on increasing PA, while the remaining studies had other objectives, such as maintaining healthy lifestyle (8 studies), weight loss management (6 studies), and rehabilitation (2 studies) as well. The reviewed studies provide personalization in six categories related to recommendation and feedback: goal recommendation, activity recommendation, fitness partner recommendation, educational content, motivational content, and intervention timing. With respect to the mode of generation, interventions were found to be semi-automated, or automatic. Of these, the automatic interventions were either knowledge-based or data-driven or both. User models in the studies were constructed with parameters from five categories: PA profile, demographics, medical data, behaviour change technique (BCT) parameters, and contextual information. Only 27 of the eligible studies evaluated the interventions for improvement in PA and 16 of these concluded that the interventions to increase PA are more effective when they are personalized. Conclusions: This scoping review investigates personalization of technology-based interventions, in the form of recommendations or feedback for increasing PA. Based on the review and gaps identified, research directions for improving the efficacy of personalized interventions are proposed. First, data-driven prediction techniques can facilitate effective personalization of automated and semi-automated interventions. Second, use of BCTs in automated interventions, and in combination with PA guidelines, are yet to be explored, and preliminary studies in this direction are promising. Third, systems with automated interventions also need to be suitably adapted to serve specific needs of patients with clinical conditions. Fourth, previous user models focus on single metric evaluations of PA instead of a, potentially more effective, holistic and multidimensional view. Fifth, with the widespread adoption of activity monitoring devices and smartphones, personalized and dynamic user models can be created using available user data, including users’ social profile. Finally, the long-term effects of such interventions, as well as the technology medium used for the interventions, need to be evaluated rigorously.
- Open Access
- 21 May 2018
- Published by JMIR Publications Inc. in JMIR Preprints
User Models for Personalized Physical Activity Interventions: Scoping Review (Preprint)
BACKGROUND Fitness devices have spurred the development of apps that aim to motivate users, through interventions, to increase their physical activity (PA). Personalization in the interventions is essential as the target users are diverse with respect to their activity levels, requirements, preferences, and behavior. OBJECTIVE This review aimed to (1) identify different kinds of personalization in interventions for promoting PA among any type of user group, (2) identify user models used for providing personalization, and (3) identify gaps in the current literature and suggest future research directions. METHODS A scoping review was undertaken by searching the databases PsycINFO, PubMed, Scopus, and Web of Science. The main inclusion criteria were (1) studies that aimed to promote PA; (2) studies that had personalization, with the intention of promoting PA through technology-based interventions; and (3) studies that described user models for personalization. RESULTS The literature search resulted in 49 eligible studies. Of these, 67% (33/49) studies focused solely on increasing PA, whereas the remaining studies had other objectives, such as maintaining healthy lifestyle (8 studies), weight loss management (6 studies), and rehabilitation (2 studies). The reviewed studies provide personalization in 6 categories: goal recommendation, activity recommendation, fitness partner recommendation, educational content, motivational content, and intervention timing. With respect to the mode of generation, interventions were found to be semiautomated or automatic. Of these, the automatic interventions were either knowledge-based or data-driven or both. User models in the studies were constructed with parameters from 5 categories: PA profile, demographics, medical data, behavior change technique (BCT) parameters, and contextual information. Only 27 of the eligible studies evaluated the interventions for improvement in PA, and 16 of these concluded that the interventions to increase PA are more effective when they are personalized. CONCLUSIONS This review investigates personalization in the form of recommendations or feedback for increasing PA. On the basis of the review and gaps identified, research directions for improving the efficacy of personalized interventions are proposed. First, data-driven prediction techniques can facilitate effective personalization. Second, use of BCTs in automated interventions, and in combination with PA guidelines, are yet to be explored, and preliminary studies in this direction are promising. Third, systems with automated interventions also need to be suitably adapted to serve specific needs of patients with clinical conditions. Fourth, previous user models focus on single metric evaluations of PA instead of a potentially more effective, holistic, and multidimensional view. Fifth, with the widespread adoption of activity monitoring devices and mobile phones, personalized and dynamic user models can be created using available user data, including users’ social profile. Finally, the long-term effects of such interventions as well as the technology medium used for the interventions need to be evaluated rigorously.
- 21 May 2018
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessUser Models for Personalized Physical Activity Interventions: Scoping Review (Preprint)
BACKGROUND Fitness devices have spurred the development of apps that aim to motivate users, through interventions, to increase their physical activity (PA). Personalization in the interventions is essential as the target users are diverse with respect to their activity levels, requirements, preferences, and behavior. OBJECTIVE This review aimed to (1) identify different kinds of personalization in interventions for promoting PA among any type of user group, (2) identify user models used for providing personalization, and (3) identify gaps in the current literature and suggest future research directions. METHODS A scoping review was undertaken by searching the databases PsycINFO, PubMed, Scopus, and Web of Science. The main inclusion criteria were (1) studies that aimed to promote PA; (2) studies that had personalization, with the intention of promoting PA through technology-based interventions; and (3) studies that described user models for personalization. RESULTS The literature search resulted in 49 eligible studies. Of these, 67% (33/49) studies focused solely on increasing PA, whereas the remaining studies had other objectives, such as maintaining healthy lifestyle (8 studies), weight loss management (6 studies), and rehabilitation (2 studies). The reviewed studies provide personalization in 6 categories: goal recommendation, activity recommendation, fitness partner recommendation, educational content, motivational content, and intervention timing. With respect to the mode of generation, interventions were found to be semiautomated or automatic. Of these, the automatic interventions were either knowledge-based or data-driven or both. User models in the studies were constructed with parameters from 5 categories: PA profile, demographics, medical data, behavior change technique (BCT) parameters, and contextual information. Only 27 of the eligible studies evaluated the interventions for improvement in PA, and 16 of these concluded that the interventions to increase PA are more effective when they are personalized. CONCLUSIONS This review investigates personalization in the form of recommendations or feedback for increasing PA. On the basis of the review and gaps identified, research directions for improving the efficacy of personalized interventions are proposed. First, data-driven prediction techniques can facilitate effective personalization. Second, use of BCTs in automated interventions, and in combination with PA guidelines, are yet to be explored, and preliminary studies in this direction are promising. Third, systems with automated interventions also need to be suitably adapted to serve specific needs of patients with clinical conditions. Fourth, previous user models focus on single metric evaluations of PA instead of a potentially more effective, holistic, and multidimensional view. Fifth, with the widespread adoption of activity monitoring devices and mobile phones, personalized and dynamic user models can be created using available user data, including users’ social profile. Finally, the long-term effects of such interventions as well as the technology medium used for the interventions need to be evaluated rigorously.