Sangjin Park
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Sangjin Park has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Jae-Suk Yang
Graduate School of Future Strategy Korea Advanced Institute of Science and Technology Daejeon Republic of Korea
5 shared publications - Hyeonhak Kim
Information Technology Development Division, National Assembly, Seoul, Republic of Korea
4 shared publications - Jinwoong Kim
Department of Industrial Data Engineering, Hanyang University, Seoul, Republic of Korea
3 shared publications - Sujung Kim2 shared publications
Hyeonhak Kim
Information Technology Development Division, National Assembly, Seoul, Republic of Korea
1 shared publicationJinwoong Kim
Department of Industrial Data Engineering, Hanyang University, Seoul, Republic of Korea
1 shared publication
Latest Publications
- Open Access
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 2138-2149
GroupSegment-SHAP: Shapley Value Explanations with Group-Segment Players for Multivariate Time Series
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 2138-2149
Open AccessGroupSegment-SHAP: Shapley Value Explanations with Group-Segment Players for Multivariate Time Series
- Open Access
- 23 September 2026
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 28, e101150
Digital Behavior Change Interventions for Digital Overuse: Scoping Review of Current Approaches and Emerging AI-Enabled Systems
Background: Digital overuse poses a significant threat to health through multiple pathways, including the deterioration of mental health and sleep disturbance. This issue has led to the development of digital overuse-targeted digital behavior change interventions (DO-DBCIs). Although research in this field has advanced by adopting state-of-the-art technologies, substantial gaps exist in elements critical to establishing intervention validity, including target devices and activities, intervention strategies, theoretical foundations, target populations, and sustainability of effects. Objective: This scoping review aimed to systematically map DO-DBCI research across these five dimensions and to identify structural gaps and future research priorities. Methods: Following PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines, 4 databases (Web of Science, ACM Digital Library, IEEE Xplore, and PubMed) were searched on August 4, 2026, for studies published through 2025. Of 3260 records retrieved, 32 studies were included after 2-stage screening by 2 independent reviewers. Results: Pronounced structural biases were identified across all 5 dimensions. Most studies targeted smartphones exclusively (27/32, 84%) and aimed at reducing total usage time without distinguishing specific activities (26/32, 81%); no study addressed short-form video platforms. Self-monitoring was the most frequently identified strategy, followed by nudge and friction, while AI- and machine learning (ML)–based interventions remained limited. Of the 32 studies, 12 (38%) lacked an explicit theoretical framework. Target populations were heavily skewed toward young adults (17/31, 55%), with only 1 of 31 studies (3%) targeting adolescents. Most studies adopted short-term designs of 4 weeks or less, and only 38% (12/32) included any postintervention follow-up. Conclusions: This first comprehensive scoping review of DO-DBCI identifies 4 critical gaps—device and activity concentration, theory-to-design disconnect, demographic skew toward convenience samples, and insufficient longitudinal evaluation. Future research should prioritize activity-specific interventions, explicit theory-driven design, expanded population diversity, and standardized longitudinal study designs.
- 23 September 2026
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 28, e101150
Open AccessDigital Behavior Change Interventions for Digital Overuse: Scoping Review of Current Approaches and Emerging AI-Enabled Systems
Background: Digital overuse poses a significant threat to health through multiple pathways, including the deterioration of mental health and sleep disturbance. This issue has led to the development of digital overuse-targeted digital behavior change interventions (DO-DBCIs). Although research in this field has advanced by adopting state-of-the-art technologies, substantial gaps exist in elements critical to establishing intervention validity, including target devices and activities, intervention strategies, theoretical foundations, target populations, and sustainability of effects. Objective: This scoping review aimed to systematically map DO-DBCI research across these five dimensions and to identify structural gaps and future research priorities. Methods: Following PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines, 4 databases (Web of Science, ACM Digital Library, IEEE Xplore, and PubMed) were searched on August 4, 2026, for studies published through 2025. Of 3260 records retrieved, 32 studies were included after 2-stage screening by 2 independent reviewers. Results: Pronounced structural biases were identified across all 5 dimensions. Most studies targeted smartphones exclusively (27/32, 84%) and aimed at reducing total usage time without distinguishing specific activities (26/32, 81%); no study addressed short-form video platforms. Self-monitoring was the most frequently identified strategy, followed by nudge and friction, while AI- and machine learning (ML)–based interventions remained limited. Of the 32 studies, 12 (38%) lacked an explicit theoretical framework. Target populations were heavily skewed toward young adults (17/31, 55%), with only 1 of 31 studies (3%) targeting adolescents. Most studies adopted short-term designs of 4 weeks or less, and only 38% (12/32) included any postintervention follow-up. Conclusions: This first comprehensive scoping review of DO-DBCI identifies 4 critical gaps—device and activity concentration, theory-to-design disconnect, demographic skew toward convenience samples, and insufficient longitudinal evaluation. Future research should prioritize activity-specific interventions, explicit theory-driven design, expanded population diversity, and standardized longitudinal study designs.
- Open Access
- 21 July 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Temporal interpretation of physiologic group contributions for ICU mortality prediction
Reliable ICU mortality prediction requires capturing physiologic evolution over time and interpreting these patterns in clinically meaningful ways. However, existing methods often focus on individual variables or isolated time points, limiting their ability to represent related physiologic states and temporal progression. We developed a group-based temporal interpretation framework by applying GroupSegment-SHAP to characterize the evolving contributions of physiologic groups before ICU mortality. Using MIMIC-IV, we evaluated prediction models across 24-, 48-, and 72-hour death-aligned lookback windows and selected Mamba, a selective state-space model, as the interpretation backbone because it showed comparatively strong and stable performance. Clinical variables were reorganized into physiologic groups, and Shapley values from the trained Mamba model were used to quantify group-wise temporal importance. Group-level interpretation identified hemodynamic, neurologic, cardiopulmonary, and temperature-related groups as major mortality predictors. Temporal analysis showed that 24-hour trajectories emphasized acute cardiopulmonary, metabolic, hematologic, and acid-base/respiratory abnormalities, whereas 48- and 72-hour trajectories additionally captured hemodynamic, renal-output, neurologic, and hepatic differences. Cardiopulmonary importance increased 8 to 30 h before peak group-level importance, when non-survivors showed lower normal-range proportions than survivors by up to 17.5% points. This analysis shifts interpretation from static variable-level importance to temporal trajectories of physiologic groups.
- 21 July 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Open AccessTemporal interpretation of physiologic group contributions for ICU mortality prediction
Reliable ICU mortality prediction requires capturing physiologic evolution over time and interpreting these patterns in clinically meaningful ways. However, existing methods often focus on individual variables or isolated time points, limiting their ability to represent related physiologic states and temporal progression. We developed a group-based temporal interpretation framework by applying GroupSegment-SHAP to characterize the evolving contributions of physiologic groups before ICU mortality. Using MIMIC-IV, we evaluated prediction models across 24-, 48-, and 72-hour death-aligned lookback windows and selected Mamba, a selective state-space model, as the interpretation backbone because it showed comparatively strong and stable performance. Clinical variables were reorganized into physiologic groups, and Shapley values from the trained Mamba model were used to quantify group-wise temporal importance. Group-level interpretation identified hemodynamic, neurologic, cardiopulmonary, and temperature-related groups as major mortality predictors. Temporal analysis showed that 24-hour trajectories emphasized acute cardiopulmonary, metabolic, hematologic, and acid-base/respiratory abnormalities, whereas 48- and 72-hour trajectories additionally captured hemodynamic, renal-output, neurologic, and hepatic differences. Cardiopulmonary importance increased 8 to 30 h before peak group-level importance, when non-survivors showed lower normal-range proportions than survivors by up to 17.5% points. This analysis shifts interpretation from static variable-level importance to temporal trajectories of physiologic groups.
- Open Access
- 12 May 2026
- Published by JMIR Publications Inc. in JMIR Preprints
Digital Behavior Change Interventions for Digital Overuse: Scoping Review of Current Approaches and Emerging AI-Enabled Systems (Preprint)
BACKGROUND Digital overuse poses a significant threat to health through multiple pathways, including the deterioration of mental health and sleep disturbance. This issue has led to the development of digital overuse-targeted digital behavior change interventions (DO-DBCIs). Although research in this field has advanced by adopting state-of-the-art technologies, substantial gaps exist in elements critical to establishing intervention validity, including target devices and activities, intervention strategies, theoretical foundations, target populations, and sustainability of effects. OBJECTIVE This scoping review aimed to systematically map DO-DBCI research across these five dimensions and to identify structural gaps and future research priorities. METHODS Following PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines, 4 databases (Web of Science, ACM Digital Library, IEEE Xplore, and PubMed) were searched on August 4, 2026, for studies published through 2025. Of 3260 records retrieved, 32 studies were included after 2-stage screening by 2 independent reviewers. RESULTS Pronounced structural biases were identified across all 5 dimensions. Most studies targeted smartphones exclusively (27/32, 84%) and aimed at reducing total usage time without distinguishing specific activities (26/32, 81%); no study addressed short-form video platforms. Self-monitoring was the most frequently identified strategy, followed by nudge and friction, while AI- and machine learning (ML)–based interventions remained limited. Of the 32 studies, 12 (38%) lacked an explicit theoretical framework. Target populations were heavily skewed toward young adults (17/31, 55%), with only 1 of 31 studies (3%) targeting adolescents. Most studies adopted short-term designs of 4 weeks or less, and only 38% (12/32) included any postintervention follow-up. CONCLUSIONS This first comprehensive scoping review of DO-DBCI identifies 4 critical gaps—device and activity concentration, theory-to-design disconnect, demographic skew toward convenience samples, and insufficient longitudinal evaluation. Future research should prioritize activity-specific interventions, explicit theory-driven design, expanded population diversity, and standardized longitudinal study designs.
- 12 May 2026
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessDigital Behavior Change Interventions for Digital Overuse: Scoping Review of Current Approaches and Emerging AI-Enabled Systems (Preprint)
BACKGROUND Digital overuse poses a significant threat to health through multiple pathways, including the deterioration of mental health and sleep disturbance. This issue has led to the development of digital overuse-targeted digital behavior change interventions (DO-DBCIs). Although research in this field has advanced by adopting state-of-the-art technologies, substantial gaps exist in elements critical to establishing intervention validity, including target devices and activities, intervention strategies, theoretical foundations, target populations, and sustainability of effects. OBJECTIVE This scoping review aimed to systematically map DO-DBCI research across these five dimensions and to identify structural gaps and future research priorities. METHODS Following PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines, 4 databases (Web of Science, ACM Digital Library, IEEE Xplore, and PubMed) were searched on August 4, 2026, for studies published through 2025. Of 3260 records retrieved, 32 studies were included after 2-stage screening by 2 independent reviewers. RESULTS Pronounced structural biases were identified across all 5 dimensions. Most studies targeted smartphones exclusively (27/32, 84%) and aimed at reducing total usage time without distinguishing specific activities (26/32, 81%); no study addressed short-form video platforms. Self-monitoring was the most frequently identified strategy, followed by nudge and friction, while AI- and machine learning (ML)–based interventions remained limited. Of the 32 studies, 12 (38%) lacked an explicit theoretical framework. Target populations were heavily skewed toward young adults (17/31, 55%), with only 1 of 31 studies (3%) targeting adolescents. Most studies adopted short-term designs of 4 weeks or less, and only 38% (12/32) included any postintervention follow-up. CONCLUSIONS This first comprehensive scoping review of DO-DBCI identifies 4 critical gaps—device and activity concentration, theory-to-design disconnect, demographic skew toward convenience samples, and insufficient longitudinal evaluation. Future research should prioritize activity-specific interventions, explicit theory-driven design, expanded population diversity, and standardized longitudinal study designs.
- Open Access
- 2 February 2026
- Published by Wiley in Corporate Social Responsibility and Environmental Management
- Vol. 33 (4) , 5236-5255
Strategic Prioritization of Environmental, Social, and Governance Goals: A Data‐Driven Efficiency Framework
Organizations are increasingly required to integrate environmental, social, and governance (ESG) objectives alongside operational performance, yet empirical guidance on how firms should prioritize among ESG activities under resource constraints remains limited. Existing research often treats ESG as a unified construct, despite differences in the mechanisms and time horizons through which environmental, social, and governance practices operate. Using longitudinal data from 2952 publicly listed firms, we combine Data Envelopment Analysis with interpretable machine learning to examine how individual ESG components are associated with firm level efficiency across different time horizons. The results indicate that social and governance dimensions are more strongly associated with efficiency in the short term, while environmental dimensions become more salient over longer horizons. The study introduces a temporal perspective on ESG efficiency and highlights the importance of timing and sequencing in ESG analysis, while recognizing that the findings reflect empirical associations rather than causal effects.
- 2 February 2026
- Published by Wiley in Corporate Social Responsibility and Environmental Management
- Vol. 33 (4) , 5236-5255
Open AccessStrategic Prioritization of Environmental, Social, and Governance Goals: A Data‐Driven Efficiency Framework
Organizations are increasingly required to integrate environmental, social, and governance (ESG) objectives alongside operational performance, yet empirical guidance on how firms should prioritize among ESG activities under resource constraints remains limited. Existing research often treats ESG as a unified construct, despite differences in the mechanisms and time horizons through which environmental, social, and governance practices operate. Using longitudinal data from 2952 publicly listed firms, we combine Data Envelopment Analysis with interpretable machine learning to examine how individual ESG components are associated with firm level efficiency across different time horizons. The results indicate that social and governance dimensions are more strongly associated with efficiency in the short term, while environmental dimensions become more salient over longer horizons. The study introduces a temporal perspective on ESG efficiency and highlights the importance of timing and sequencing in ESG analysis, while recognizing that the findings reflect empirical associations rather than causal effects.
- Open Access
- 1 January 2026
- Published by Wiley in International Journal of Intelligent Systems
- Vol. 2026 (1)
International Stock Forecasting Using Ensemble Deep Graph Models and Complex Network Analysis
International stock forecasting faces challenges as global stock markets become increasingly synchronized. This study develops a prediction model for 46 global stock prices by examining the complex interconnectedness of the global stock network. We propose a multilevel fusion approach that integrates technology and knowledge for accurate international stock predictions. First, for technical convergence, we introduce a hybrid model combining graph convolutional networks (GCNs) and long short‐term memory (LSTM) to capture both temporal dynamics and spatial dependencies. Additionally, we construct an ensemble of multiple GCN‐LSTM models via AdaBoost, which reduces mean absolute error (MAE) by an average of 36.7% compared with baseline models. Second, for knowledge fusion, we construct stock centrality features that represent influence on the entire network market for each country and use this feature as input to models, further decreasing MAE by 23.2%. In profit tests, trading based on our predictions significantly improved cumulative profit and the Sharpe ratio by 2.06× and 2.83×, respectively, compared with a simple long‐holding strategy. Finally, using Shapley Additive Explanations (SHAP), we derived the insight that our stock network feature has a stronger impact on future stock prices during high‐volatility periods, such as when the financial system experiences shocks. These findings suggest that the network index becomes particularly valuable in enhancing predictive performance under turbulent market conditions.
- 1 January 2026
- Published by Wiley in International Journal of Intelligent Systems
- Vol. 2026 (1)
Open AccessInternational Stock Forecasting Using Ensemble Deep Graph Models and Complex Network Analysis
International stock forecasting faces challenges as global stock markets become increasingly synchronized. This study develops a prediction model for 46 global stock prices by examining the complex interconnectedness of the global stock network. We propose a multilevel fusion approach that integrates technology and knowledge for accurate international stock predictions. First, for technical convergence, we introduce a hybrid model combining graph convolutional networks (GCNs) and long short‐term memory (LSTM) to capture both temporal dynamics and spatial dependencies. Additionally, we construct an ensemble of multiple GCN‐LSTM models via AdaBoost, which reduces mean absolute error (MAE) by an average of 36.7% compared with baseline models. Second, for knowledge fusion, we construct stock centrality features that represent influence on the entire network market for each country and use this feature as input to models, further decreasing MAE by 23.2%. In profit tests, trading based on our predictions significantly improved cumulative profit and the Sharpe ratio by 2.06× and 2.83×, respectively, compared with a simple long‐holding strategy. Finally, using Shapley Additive Explanations (SHAP), we derived the insight that our stock network feature has a stronger impact on future stock prices during high‐volatility periods, such as when the financial system experiences shocks. These findings suggest that the network index becomes particularly valuable in enhancing predictive performance under turbulent market conditions.
- Open Access
- 8 March 2025
- Published by Springer Nature in International Journal of Information Security
- Vol. 24 (2) , 1
Correction: Cryptocurrency recovery framework using pre-signed transaction
- 8 March 2025
- Published by Springer Nature in International Journal of Information Security
- Vol. 24 (2) , 1
Open AccessCorrection: Cryptocurrency recovery framework using pre-signed transaction
- 22 February 2025
- Published by Springer Nature in International Journal of Information Security
- Vol. 24 (2) , 1-16
Cryptocurrency recovery framework using pre-signed transaction
- 22 February 2025
- Published by Springer Nature in International Journal of Information Security
- Vol. 24 (2) , 1-16
Cryptocurrency recovery framework using pre-signed transaction
- 1 July 2024
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 135
Machine learning models based on bubble analysis for Bitcoin market crash prediction
- 1 July 2024
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 135
Machine learning models based on bubble analysis for Bitcoin market crash prediction
- 20 September 2023
- Published by Elsevier in Expert Systems with Applications
- Vol. 237
Machine learning modeling to forecast uncertainty between capital sudden stop and boom
- 20 September 2023
- Published by Elsevier in Expert Systems with Applications
- Vol. 237
Machine learning modeling to forecast uncertainty between capital sudden stop and boom