Data-driven Perspective Research Articles

A hybrid mechanism-based and data-driven model for efficient indoor temperature distribution prediction with transfer learning

Efficient prediction of indoor temperature distribution (ITD) is crucial for various building-related applications, particularly in real-time thermal management. Despite efforts from both physics-based and data-driven perspectives, ITD prediction still faces challenges, particularly in balancing prediction accuracy with computational speed and addressing the need for extensive high-quality data in practical applications. To address these challenges, this study develops a hybrid temperature distribution prediction model (HTDPM) that integrates deep learning with physical mechanisms, enabling the effective capture of spatial-temporal dependencies and uncertainties with limited input parameters, achieving a balance between prediction accuracy and computational efficiency. Meanwhile, a novel transfer learning approach is applied to HTDPM (HTDPM-TL), significantly reducing data requirements and enhancing model generalization across various practical scenarios. Besides, orthogonal and randomized experiments are designed to simulate multiple real-world scenarios, thus constructing an extensive source domain dataset. The HTDPM-TL was tested in various real-world scenarios with several baseline methods, demonstrating an average RMSE of 0.794 ∘C, a computational time of 0.289 s, and limited data volume. The computational speed was improved by three orders of magnitude compared to CFD simulations, and the prediction resolution was enhanced threefold compared to traditional data-driven models. These results highlight the potential of HTDPM-TL to achieve an optimal trade-off between prediction accuracy, computational efficiency, and data requirement.

Bibliometric Insights into Mental Health Challenges and Innovations in Contemporary Society

This bibliometric analysis examines the evolution and current state of mental health research from a comprehensive, data-driven perspective. Using VOSviewer for visual mapping, this study highlights key terms, thematic developments, and influential authors within the field over several decades. The analysis reveals a strong ongoing focus on core issues such as "mental illness," "anxiety," and "depression," alongside a notable rise in the integration of technological innovations like telepsychiatry and mobile health applications. The study also identifies significant authorship networks that suggest robust collaborative efforts, which are crucial for advancing research and implementing effective mental health interventions. The findings point towards an increasing recognition of the impact of social and environmental factors on mental health, suggesting new directions for future research and policy-making. This study underscores the need for a dynamic approach to mental health research, one that adapts to emerging challenges and leverages technological advances to improve access to care and treatment outcomes.

Multi-level attention graph feature fusion smooth prognostics approach for aircraft engines remaining useful life prediction

In the development of a prognostic and health management system, it is a challenge to mine the degradation characteristics, estimate the RUL reliably and smoothly, and explain the RUL prediction results from the multi-sensor condition monitoring data collected under the non-fixed-length equipment operation cycle and multi-operating conditions. In this paper, a Multi-level Attention Graph feature Fusion Smooth Prognostics Approach (MAGF-SPA) is proposed to achieve accurate, smooth, and interpretable RUL prediction results. In this method, firstly, multivariate time series (MTS) are transformed into graph signals from a purely data-driven perspective. Secondly, the four-level feature extraction strategy, which includes node-level, subgraph-level, cyclical-level, and temporal-level feature extraction, is designed to extract the inherent hidden degradation characteristics of MTS. Finally, a smoothing trick proposed in this paper can optimise the RUL prediction value corresponding to the degradation characteristics in real time. Experimental results show that the proposed method achieves state-of-the-art on the N-CMAPSS dataset.

Power flow analysis and volt/var control strategy of the active distribution network based on data-driven method

In order to solve the problems of low power flow calculation accuracy and voltage overcrossing of the distribution network caused by large-scale distributed power supply, a data-driven power flow analysis and volt/var optimization control strategy for distribution network are proposed in this paper. Firstly, a data-driven power flow analysis model for the distribution network is proposed, and the nonlinear mapping relationship between the distribution network state and power flow results is described from the data-driven perspective. Secondly, the paper analyzes the influence of photovoltaic power supply on the voltage of the distribution network, and establishes the volt/var optimization model based on photovoltaic power supply. Then, on the basis of data-driven power flow analysis of the distribution network, the distribution network reactive voltage optimization strategy based on data-driven power flow analysis is proposed to ensure the safe and stable operation of distribution network voltage, reduce the operating network loss of distribution network, and realize the distribution network reactive voltage optimization is not affected by factors such as distribution network line parameters. Finally, IEEE 33 node power distribution system is used to verify the effectiveness of the proposed strategy.

A Study on the Layoff Problem After COVID-19 Based on Decision Tree Ensemble Methods

The COVID-19 pandemic, a cataclysmic event in modern history, has upended the global economy, precipitating extensive layoffs across diverse sectors. This study delves into the determinants of these workforce reductions through the lens of decision tree ensemble methods. By scrutinizing a comprehensive dataset, we harnessed statistical and machine learning algorithms to assess the significance of key variablesincluding industry type, geographic location, company development stage, and capital acquisitionon the scale of layoffs. This analysis further entailed predicting the total number of employees laid off and examining their distribution. The use of decision tree ensemble methods, such as random forests and gradient boosting, provided robust insights into the complex interplay of factors influencing layoff decisions. We discovered that industry type and company development stage were particularly critical in predicting layoff patterns, while geographic location and capital acquisition also played notable roles. This research offers a data-driven perspective on the layoff phenomena, shedding light on the multifaceted influences at play and offering a foundation for further inquiry and strategic policy development in response to economic downturns. By understanding these dynamics, policymakers and business leaders can better navigate future economic crises, potentially mitigating the adverse impacts on the workforce and promoting more resilient economic structures.

Systematic analysis and network mapping of disease associations in autoimmune polyglandular syndrome.

The purpose of our work was to provide a data-driven perspective to APS, a complex autoimmune disorder, supplementing traditional clinical observations. Medical records of 7559 patients were analyzed, autoimmune origin was proved in 3180 cases of which 380 (12%) had APS. Associations of component disorders were investigated by computational methods to reveal typical patterns of disease development. Twenty-eight distinct autoimmune disorders were diagnosed forming 113 combinations. The 10 most frequent combinations were responsible for 51,3% of cases. HT and GD were differentiated as main cornerstones of APS, sharing several comorbidities. HT was the most common manifestation (67.4%), followed by GD (26.8%) and T1D (20.8%). APS started significantly earlier in men than in women. Thyroid autoimmunity was frequently linked to gastrointestinal and systemic manifestations and these patterns of associations substantially differed from that of T1D, AD or CeD when present as first manifestations, suggesting the possibility of a common biological cause. APS is more frequent than reported. Classifying APS requires a shift of perspective towards disease associations rather than disorder prevalence.

Intelligent monitoring of marine vessel dynamics based on data mining

This study explores vessel navigation's universal characteristics and individual anomaly behavior from dispatch to destination. With the help of information technology, it introduces big data and machine learning technologies into the work of vessel traffic management and conducts statistical analysis and monitoring studies of vessel behavior from a data-driven perspective. The study was conducted by modeling vessel speed, position, course, and other navigational characteristics behavior through comprehensive statistical analysis of vessel AIS, radar, and VTS data. The study was based on the monitoring methods for anomalous behavior of vessel speed, position, and heading inconsistencies with the expected behavior identified by the model.

The Role of E-Government in Public Services: A Bibliometric Analysis

This research investigates the pivotal role of e-government in public services through the lens of bibliometric analysis, employing the Biblioshiny tool for comprehensive bibliometric evaluation. E-government, a transformative force in contemporary public administration, involves the use of electronic platforms to deliver government services to citizens. The study employs bibliometric analysis to systematically examine and quantify the scholarly output, trends, and patterns in research related to the role of e-government in public services. The research leverages Biblioshiny, a sophisticated bibliometric analysis tool, to conduct an in-depth examination of a diverse range of publications. This tool facilitates the extraction of valuable insights from academic literature, shedding light on key thematic areas, prolific authors, and publication trends over specified periods. By employing bibliometric analysis, the study aims to provide a comprehensive overview of the intellectual landscape surrounding e-government in public services. The findings of this research are expected to contribute significantly to the understanding of the scholarly discourse on the role of e-government in public services. By utilizing bibliometric analysis, the study aims to identify gaps in existing research, highlight areas of prolific academic activity, and provide insights that can inform future research directions and policy considerations in the realm of e-government. Ultimately, this research endeavours to offer a nuanced and data-driven perspective on the evolving landscape of e-government and its impact on shaping effective and citizen-centric public services, thereby enhancing the efficiency, transparency, and accessibility of government operations and interactions with the public.

Enhancing transparency in data-driven urban pluvial flood prediction using an explainable CNN model

Mitigating severe losses caused by pluvial floods in urban areas with dense population and property requires effective flood prediction for emergency measures. Physics-based models face issues with low computational efficiency for real‐time flood prediction. An alternative approach for rapid prediction instead of physics-based models is to predict from a data-driven perspective. However, data-driven approaches for urban flood prediction are often perceived as “black box” and might raise concerns. In this study, we propose an explainable deep learning (DL) approach for rapid urban pluvial flood prediction with enhanced transparency using a convolutional neural network (CNN) and the explainable artificial intelligence (AI) framework Shapley additive explanation (SHAP). We process a systematic stepwise feature selection process and establish a CNN model considering topography, drainage networks and rainfall to predict maximum inundation depths. Then, SHAP is applied to provide trustworthy explanations for the decision making in model results. The results show that: 1) Forward selection can offer insights into selecting effective input variables for improved predictions and promote understanding of DL modelling. The spatial pattern of inundation depths predicted by the proposed CNN model shows good agreement with those predicted by the physics-based model, demonstrated by average correlation coefficient (CC) and mean absolute error (MAE) values of 0.982 and 0.021 m, respectively. 2) The CNN model substantially outperforms the physics-based model in computational speed when using the same hardware, achieving speedups of 210 times on GPU and 51 times on CPU in the case study (575167 grid cells, 14.38 km2). Particularly, it can still achieve good performance on a CPU-only standard laptop without high-performance hardware, with only a modest increase in computational time. 3) The SHAP explainable analysis confirms that the CNN model correctly captures the relationships between input variables and water depth, revealing a reasonable decision-making process, enhancing its transparency. The explainable DL approach incorporating SHAP for rapid urban pluvial flood prediction is promising to build trust among stakeholders and provide a trustworthy reference for prompt measures aiming at saving lives and protecting assets during flood emergencies. Additionally, the proposed DL approach can potentially be further expanded to analyze the causes of urban flooding events and serve as a foundation for exploring the transferability of data-driven urban flood prediction, providing benefits for better urban flood risk management.

IFNet: Data-driven multisensor estimate fusion with unknown correlation in sensor measurement noises

In recent years, multisensor fusion for state estimation has gained considerable attention. The effectiveness of the optimal fusion estimation method heavily relies on the correlation among sensor measurement noises. To enhance estimate fusion performance by mining unknown correlation in the data, this paper introduces a novel multisensor fusion approach using an information filtering neural network (IFNet) for discrete-time nonlinear state space models with cross-correlated measurement noises. The method presents three notable advantages: First, it offers a data-driven perspective to tackle uncertain correlation in multisensor estimate fusion while preserving the interpretability of the information filtering. Second, by harnessing the RNN’s capability to manage data streams, it can dynamically update the fusion weights between sensors to improve fusion accuracy. Third, this method has a lower complexity than the state-of-the-art KalmanNet measurement fusion method when dealing with the fusion problem involving a large number of sensors. Numerical simulations demonstrate that IFNet exhibits better fusion accuracy than traditional filtering methods and KalmanNet fusion filtering when correlation among measurement noises is unknown.

Exploring Pedagogical Dynamics via Teachers’ Interactions with Digital Textbook Platforms

Understanding the pedagogical dynamics that occur when teachers interact with digital textbook devices (DTDs) in the artificial intelligence (AI)–enhanced teaching environment is critical for education success. Previous research on the use of DTDs has primarily focused on subjective and intrusive questionnaire surveys, with little attention paid to the accompanying data when teachers use AI tools. This study aims to mine teachers’ behaviors from a data-driven perspective, exploring actual practices as they integrate DTDs into their teaching. We collected teachers’ behavioral data from a cloud-based curriculum platform, covering four subjects over a four-month period. Then, we used the lag sequential analysis (LSA) technique to investigate state-change patterns and the motif discovery technique to uncover hidden sequential patterns in teachers’ behaviors. Furthermore, using two mutually dependent recurrent neural networks, we predicted teachers’ future behaviors while operating DTDs, providing a more complete understanding of their pedagogical dynamics. Our findings show significant behavioral patterns when teachers use AI–embedded digital devices in the smart teaching environment. Additionally, the predictive model performed well in predicting teachers’ future behaviors, which can be used to improve the design of human–computer interaction in digital textbooks.

A Review on Lithium-Ion Battery Modeling from Mechanism-Based and Data-Driven Perspectives

As the low-carbon economy continues to advance, New Energy Vehicles (NEVs) have risen to prominence in the automotive industry. The design and utilization of lithium-ion batteries (LIBs), which are core component of NEVs, are directly related to the safety and range performance of electric vehicles. The requirements for a refined design of lithium-ion battery electrode structures and the intelligent adjustment of charging modes have attracted extensive research from both academia and industry. LIB models can be divided into mechanism-based models and data-driven models; however, the distinctions and connections between these two kinds of models have not been systematically reviewed as yet. Therefore, this work provides an overview and perspectives on LIB modeling from both mechanism-based and data-driven perspectives. Meanwhile, the potential fusion modeling frameworks including mechanism information and a data-driven method are also summarized. An introduction to LIB modeling technologies is presented, along with the current challenges and opportunities. From the mechanism-based perspective of LIB structure design, we further explore how electrode morphology and aging-related side reactions impact battery performance. Furthermore, within the realm of battery operation, the utilization of data-driven models that leverage machine learning techniques to estimate battery health status is investigated. The bottlenecks for the design, state estimation, and operational optimization of LIBs and potential prospects for mechanism-data hybrid modeling are highlighted at the end. This work is expected to assist researchers and engineers in uncovering the potential value of mechanism information and operation data, thereby facilitating the intelligent transformation of the lithium-ion battery industry towards energy conservation and efficiency enhancement.

On particle dispersion statistics using unsupervised learning and Gaussian mixture models

Understanding the dispersion of particles in enclosed spaces is crucial for controlling the spread of infectious diseases. This study introduces an innovative approach that combines an unsupervised learning algorithm with a Gaussian mixture model to analyze the behavior of saliva droplets emitted from a coughing individual. The algorithm effectively clusters data, while the Gaussian mixture model captures the distribution of these clusters, revealing underlying sub-populations and variations in particle dispersion. Using computational fluid dynamics simulation data, this integrated method offers a robust, data-driven perspective on particle dynamics, unveiling intricate patterns and probabilistic distributions previously unattainable. The combined approach significantly enhances the accuracy and interpretability of predictions, providing valuable insights for public health strategies to prevent virus transmission in indoor environments. The practical implications of this study are profound, as it demonstrates the potential of advanced unsupervised learning techniques in addressing complex biomedical and engineering challenges and underscores the importance of coupling sophisticated algorithms with statistical models for comprehensive data analysis. The potential impact of these findings on public health strategies is significant, highlighting the relevance of this research to real-world applications.

The Challenges of Implementing Cognitive Computing in Small Construction Projects: A Data-Driven Perspective

This study aims to identify and analyze the challenges of implementing cognitive computing in small construction projects, where decision-making, process optimization, and sustainability enhancements are crucial yet challenging. The research adopts a mixed-methods approach, integrating a thorough literature review, quantitative evaluation, and structural equation modeling (SEM) to explore the relationships between the identified barriers and the effective application of cognitive computing. The findings reveal significant hurdles, including complexity in customization (β = 0.327, t = 9.848, p < 0.001), data integrity and integration issues (β = 0.389, t = 14.534, p < 0.001), financial and cultural constraints (β = 0.295, t = 7.850, p < 0.001), and ethical and privacy concerns (β = 0.319, t = 8.963, p < 0.001). These barriers impede the seamless adoption of cognitive computing technologies. This research contributes novel insights into the specific challenges faced by small construction projects and provides practical recommendations to overcome these obstacles. By addressing these challenges, this study offers valuable guidance for stakeholders aiming to leverage cognitive computing to improve project outcomes in the construction industry. The novelty of this research lies in its focus on small-scale projects, a relatively underexplored area, and its comprehensive analysis of the multifaceted barriers that hinder the successful implementation of cognitive computing. Doi: 10.28991/CEJ-2024-010-09-011 Full Text: PDF

Developing a machine learning-based instrument for subjective well-being assessment on Weibo and its psychological significance: An evaluative and interpretive research.

Demystifying machine learning (ML) approaches through the synergy of psychology and artificial intelligence can achieve a balance between predictive and explanatory power in model development while enhancing rigor in validation and reporting standards. Accordingly, this study aimed to bridge this research gap by developing a subjective well-being (SWB) prediction model on Weibo, serving as a psychological assessment instrument and explaining the model construction based on psychological knowledge. The model establishment involved the collection of SWB scores and posts from 1,427 valid Weibo users. Multiple machine learning algorithms were employed to train the model and fine-tune its parameters. The optimal model was selected by comparing its criterion validity and split-half reliability performance. Furthermore, SHAP values were calculated to rank the importance of features, which were then used for model interpretation. The criterion validity for the three dimensions of SWB ranged from 0.50 to 0.52 (P< 0.001), and the split-half reliability ranged from 0.94 to 0.96 (P< 0.001). The identified relevant features were related to four main aspects: cultural values, emotions, morality, and time and space. This study expands the application scope of SWB-related psychological theories from a data-driven perspective and provides a theoretical reference for further well-being prediction.

Using composite indicators and city dashboards to promote place-based policy interventions

Spatial inequality is a common urban phenomena in cities around the world, where stark contrasts in a variety of different social and economic outcomes paint a vivid picture of compound inequalities. Tackling these influences from a policy perspective remains challenging, as political economies often span multiple actors and municipal bodies, lacking effective policy instruments to challenge multiple forms of inequality at once. This paper provides a new data-driven perspective, which seeks to improve how policy is developed when trying to mitigate the impacts of compound inequality. Utilising a place-based approach, we present an evidence base which has been co-produced with policymakers, comprising composite spatial indicators and a city dashboard for Liverpool City Region. The assembled evidence base highlights clear patterns of compound inequality across the region, identifying places in greatest need of support. In the paper we discuss how this evidence base is now being used to distribute investment from the City Region Sustainable Transport Settlements, generating positive outcomes for people and places across the region. Finally, we conclude by reflecting on the benefits of building collaborative relationships between academics and policymakers, and the utility of our approach, which uses urban indicators and city dashboards, which we argue can secure a more equitable future for cities globally.

Machine Learning Predicts Peripherally Inserted Central Catheters-Related Deep Vein Thrombosis Using Patient Features and Catheterization Technology Features.

This study aims to use patient feature and catheterization technology feature variables to train the corresponding machine learning (ML) models to predict peripherally inserted central catheters-deep vein thrombosis (PICCs-DVT) and analyze the importance of the two types of features to PICCs-DVT from the aspect of "input-output" correlation. To comprehensively and systematically summarize the variables used to describe patient features and catheterization technical features, this study combined 18 literature involving the two types of features in predicting PICCs-DVT. A total of 21 variables used to describe the two types of features were summarized, and feature values were extracted from the data of 1,065 PICCs patients from January 1, 2021 to August 31, 2022, to construct a data sample set. Then, 70% of the sample set is used for model training and hyperparameter optimization, and 30% of the sample set is used for PICCs-DVT prediction and feature importance analysis of three common ML classification models (i.e. support vector classifier [SVC], random forest [RF], and artificial neural network [ANN]). In terms of prediction performance, this study selected four metrics to evaluate the prediction performance of the model: precision (P), recall (R), accuracy (ACC), and area under the curve (AUC). In terms of feature importance analysis, this study chooses a single feature analysis method based on the "input-output" sensitivity principle-Permutation Importance. For the mean model performance, the three ML models on the test set are P = 0.92, R = 0.95, ACC = 0.88, and AUC = 0.81. Specifically, the RF model is P = 0.95, R = 0.96, ACC = 0.92, AUC = 0.86; the ANN model is P = 0.92, R = 0.95, ACC = 0.88, AUC = 0.81; the SVC model is P = 0.88, R = 0.94, ACC = 0.85, AUC = 0.77. For feature importance analysis, Catheter-to-vein rate (RF: 91.55%, ANN: 82.25%, SVC: 87.71%), Zubrod-ECOG-WHO score (RF: 66.35%, ANN: 82.25%, SVC: 44.35%), and insertion attempt (RF: 44.35%, ANN: 37.65%, SVC: 65.80%) all occupy the top three in the ML models prediction task of PICCs-DVT, showing relatively consistent ranking results. The ML models show good performance in predicting PICCs-DVT and reveal a relatively consistent ranking of feature importance from the data. The important features revealed might help clinical medical staff to better understand and analyze the formation mechanism of PICCs-DVT from a data-driven perspective.

Predicting vessel service time: A data-driven approach

Vessel Service Time (VST) refers to the period from when a ship arrives at a berth until it departs. VST is a critical metric for port operational efficiency and service quality. Uncertainty in VST can undermine the operational efficiency in port management and lead to financial setbacks. To mitigate this uncertainty and lay the foundation for subsequent berth allocation, vessels typically provide an estimated departure time (EDT). However, substantial discrepancies often exist between the reported EDT and the actual departure time (ADT). These discrepancies mainly stem from unforeseen port handling inefficiencies and supply chain disruptions. This variability results in significant differences between the actual VST and its anticipated duration, thereby complicating port operations. To tackle this issue, our research represents the first study to predict VST from a data-driven perspective. We introduce an advanced tree-based stacking regression model for VST prediction, utilizing vessel port call records from 2020 to 2023. Our machine learning stacking approach achieves more accurate VST predictions than EDT reported by vessels, significantly reducing the mean absolute error (MAE) by 29.7% (from 4.54 to 3.19 h) and the root mean square error (RMSE) by 31.9% (from 6.58 to 4.48 h). The model also demonstrates reliable predictive power with an R-squared (R2) value of 0.8. These results underscore the significant scientific value of data-driven approaches in maritime studies. Our findings highlight the potential of the proposed tree-based models to surpass traditional models and originally reported data in predictive accuracy for VST. This advancement not only represents a notable improvement in predictive capabilities for VST but also lays the groundwork for further research into enhancing vessel scheduling efficiency through machine learning.

Tensions between institutional and professional frames in team talk in gerontological social work

Team meetings are central to social workers’ decision-making practices. These meetings often function as a forum for collegial consultations, when applications are processed and recommendations on decisions are discussed. In this paper, we present findings from a case study on team talk and decision-making practices in gerontological social work. The data come from a body of material gathered within the framework of a larger project covering the process of assessing elder care for older persons in three Swedish municipalities. The case concerns an application, due to homelessness, from a couple for an apartment in special housing. The team meeting was analysed using a data-driven perspective within a micro-analytical approach to talk, focusing in detail on how conflicting perspectives in the assessment of the couple’s needs are dealt with, and how tensions between divergent views and opinions are handled in relation to institutional and professional conversational frames. The findings show how the care managers (in Sweden the professional title for social workers working in elder care) negotiated the boundaries of responsibility and power within both the institutional and professional frames, revealing that the institutional frame dominated when it came to making decisions. The findings have implications for practice, as they give insight into the interactional dynamics involved in social workers’ assessments when navigating different conversational frames within their decision-making practices.

Borrowing Historical Information Across All Concentration Groups with an Application of Potency Estimation in Aquatic Toxicity Assessment

AbstractAquatic toxicity tests assess the negative impact of toxicants on survival, reproduction, and growth of organisms. Since the tests are often repeated in the same laboratory, borrowing historical experimental outcomes that are congruent would increase the precision of potency estimation with current data. In this paper, we extend the historical borrowing from control only to historical data from all concentration levels in toxicity assessments with current data using the calibrated power prior. We assume the transportability (i.e., identical concentration–response relationship) between historical and current experimental outcomes while penalizing the historical data borrowing in each concentration group independently with a modified distribution-informed calibration process. This modified calibration process prevents the over/underfitting issues by sufficiently exploring the distribution of congruence level and allows researchers to determine the level of historical information borrowing, either conservatively or aggressively, from a data-driven perspective. The effectiveness and flexibility of our proposed method are demonstrated via simulation studies. As shown in the simulation study, in scenarios when the transportability assumption fails, the proposed method still tremendously aids in the precision of toxicity assessment. The proposed method is applied in a $$\textit{Ceriodaphnia dubia}$$ Ceriodaphnia dubia test.