Data-informed Decision-making Research Articles

A value-of-information-based optimal sensor placement design framework for cost-effective structural health monitoring (with application to miter gate monitoring)

Structural health monitoring (SHM) involves collecting information to assess the health of a structure, typically to guide risk-informed maintenance decision-making or predict limit state behavior throughout its lifespan. Although the value of information (VoI) obtained from an SHM system can facilitate improved decision-making, it is important to estimate its overall utility by considering the costs involved in designing, developing, installing, maintaining, and operating the system. A feasible SHM system provides greater expected returns resulting from data-informed lifecycle management decisions than the cost of the SHM system design, fabrication, deployment, operation, and maintenance. That is, since data acquisition is a precursor to data-informed decision-making, the design of an SHM system governs its feasibility. Such cost-benefit analyses are a current topic of research in the SHM community. One approach that has been proposed for these analyses is a preposterior decision analysis using the VoI metric. In this paper, we propose a sensor optimization framework that maximizes the VoI over the structure’s lifecycle, constrained by a pre-decided maintenance policy. We use two different VoI metrics: the traditional expected VoI and a gambling-theory-inspired normalized expected-savings-to-investmentrisk ratio. We introduce three time-normalized, unitless VoI metrics that are valuable for evaluating the performance of an SHM design over an extended period. Furthermore, we consider the effect of different risk profiles on the overall optimal sensor design, recognizing that the cost of decision-making depends on the utility and risk perception of the decision-maker. We conduct a detailed analysis on the marginal utility gain of gathered information as we add more sensors, observing the utility to diminish in line with the law of diminishing returns as the information content increases. This framework is applied to the design of an SHM system used for monitoring miter gates.

An Analysis of SRSS-IE Externalizing Cut Scores to Facilitate Data-Informed Decision-Making in K–12 Schools

We report the results of a convergent validity study examining the externalizing subscale (SRSS-E5, five items) of the adapted Student Risk Screening Scale for Internalizing and Externalizing (SRSS-IE 9) with the externalizing subscale of the Teacher Report Form (TRF) with two samples of K–12 students. Results of logistic regression and receiver operating characteristic curves analyzed with a sample of 195 K–5 students from 13 schools across three states suggested the following preliminary cutting scores for the SRSS-E5 for use at the elementary level: 0 to 1 for low risk, 2 to 3 for moderate risk, and 4 to 15 for high risk for externalizing behaviors. Results of the same analyses with a sample of 227 students in Grades 6 to 12 suggested the following preliminary cutting scores for the SRSS-E5 for use in middle and high (secondary) schools: 0 to 4 for low risk, 5 to 6 for moderate risk, and 7 to 15 for high risk for externalizing behaviors. For both samples, sensitivity and specificity levels were acceptable, as was the correct classification rate. We discuss findings and limitations and emphasize the importance of replication before shifting screening practices from the SRSS-IE 12 (12 items) to the newly adapted SRSS-IE 9 (9 items).

IoT—A Promising Solution to Energy Management in Smart Buildings: A Systematic Review, Applications, Barriers, and Future Scope

The use of Internet of Things (IoT) technology is crucial for improving energy efficiency in smart buildings, which could minimize global energy consumption and greenhouse gas emissions. IoT applications use numerous sensors to integrate diverse building systems, facilitating intelligent operations, real-time monitoring, and data-informed decision-making. This critical analysis of the features and adoption frameworks of IoT in smart buildings carefully investigates various applications that enhance energy management, operational efficiency, and occupant comfort. Research indicates that IoT technology may decrease energy consumption by as much as 30% and operating expenses by 20%. This paper provides a comprehensive review of significant obstacles to the use of IoT in smart buildings, including substantial initial expenditures (averaging 15% of project budgets), data security issues, and the complexity of system integration. Recommendations are offered to tackle these difficulties, emphasizing the need for established processes and improved coordination across stakeholders. The insights provided seek to influence future research initiatives and direct the academic community in construction engineering and management about the appropriate use of IoT technology in smart buildings. This study is a significant resource for academics and practitioners aiming to enhance the development and implementation of IoT solutions in the construction sector.

Enhancing Education and Teaching Management Through Data Mining and Support Vector Machine Algorithm

This study investigates the effectiveness of the SVM+TLBO approach for predicting student performance and evaluating learning abilities in educational settings. By integrating Support Vector Machines (SVM) with Teaching-Learning-Based Optimization (TLBO), the research aims to enhance predictive accuracy and efficiency compared to traditional methods, including Decision Trees, Ant Colony Algorithms, Clustering Algorithms, Convolutional Neural Networks (CNN), Neural Networks, Support Vector Machine (SVM) without Optimization. Results indicate that the SVM+TLBO model significantly outperforms these methods, providing robust predictions and valuable insights into student learning patterns. The findings underscore the importance of data-informed decision-making in education, enabling educators to identify at-risk students and tailor instructional strategies effectively. This research contributes to the growing field of educational data mining, highlighting the potential of advanced machine learning techniques to optimize educational outcomes and foster personalized learning experiences.

Equipping Preservice Teachers for Data Use: A Study of Secondary Educator Preparation Programs in Virginia

Because data-informed decision-making (DIDM) can help teachers meet diverse learners’ needs (van Geel et al., 2016), educator preparation programs (EPPs) must ensure that preservice teachers (PSTs) develop the data literacy skills needed for effective data use. However, little is known about the ways in which EPPs work towards building PSTs’ data literacy, despite licensure and accreditation requirements compelling EPPs to do so. In this study, we analyzed survey, document, and interview data from Virginia EPPs to determine what present practices for DIDM preparation are taking place across the state. Results point to a lack of uniformity among EPPs for how preparation is undertaken, and that PSTS seem to have limited coursework on data use. Additionally, there appears to be minimal collaboration between EPPs and clinical partners, such that PSTs infrequently have opportunities to engage in DIDM during field experiences.

Revolutionizing Education: The Impact of AI-Enhanced Teaching Strategies

In recent years, the integration of Artificial Intelligence (AI) into educational settings has revolutionized traditional teaching methodologies, leading to the development of innovative teaching strategies. This paper explores how AI-enhanced approaches are transforming education by personalizing learning experiences, improving student engagement, and optimizing instructional methods. Through adaptive learning platforms, AI tailors educational content to meet individual student needs, fostering a more inclusive and effective learning environment. Intelligent tutoring systems provide real-time feedback and support, enabling students to progress at their own pace and achieve mastery in various subjects. Additionally, AI-driven analytics offer educators valuable insights into student performance and learning patterns, facilitating data-informed decisionmaking and targeted interventions. By leveraging AI technologies, educators can create dynamic, interactive, and studentcentered classrooms that cater to diverse learning styles and preferences. This paper discusses the potential benefits and challenges of AI-enhanced teaching strategies and highlights the importance of integrating AI thoughtfully and ethically to maximize its positive impact on education.

Constructing a Predictive Model for High School Students' Enrolment Results

Nowadays, data-informed decision-making can assist high-school educational decision-making, especially student test scores, and further education data, then these data can be used to inform decision-making in schools and boost overall school performance. We aimed to assess the relationships among high school students' admission scores, academic performance at school, and psychological tests in the school system. The relationships were used to predict the departmental fields and universities that students would select for advanced studies in the future. This research will utilize data mining algorithms, including Naïve Bayes, SVM, and Neural Network classifiers, to extract and analyze hidden patterns in students' academic records and credentials. When the Bayesian Network model was used to predict whether a student's university is public or private, our results showed 78.46% of the predicted results. Predicting the accuracy of the top three departmental fields, the SVM model achieved the highest performance with an accuracy of 69.77% .Given that students' academic performance varies across high schools, each school should develop its customized predictive model using only its students’ data. This will help students believe in and pursue their interests.

Leveraging deep learning and computer vision technologies to enhance management of coastal fisheries in the Pacific region

This paper presents the design and development of a coastal fisheries monitoring system that harnesses artificial intelligence technologies. Application of the system across the Pacific region promises to revolutionize coastal fisheries management. The program is built on a centralized, cloud-based monitoring system to automate data extraction and analysis processes. The system leverages YoloV4, OpenCV, and ResNet101 to extract information from images of fish and invertebrates collected as part of in-country monitoring programs overseen by national fisheries authorities. As of December 2023, the system has facilitated automated identification of over six hundred nearshore finfish species, and automated length and weight measurements of more than 80,000 specimens across the Pacific. The system integrates other key fisheries monitoring data such as catch rates, fishing locations and habitats, volumes, pricing, and market characteristics. The collection of these metrics supports much needed rapid fishery assessments. The system’s co-development with national fisheries authorities and the geographic extent of its application enables capacity development and broader local inclusion of fishing communities in fisheries management. In doing so, the system empowers fishers to work with fisheries authorities to enable data-informed decision-making for more effective adaptive fisheries management. The system overcomes historically entrenched technical and financial barriers in fisheries management in many Pacific island communities.

AI-Augmented LangChain: Facilitating Natural Language SQL Queries for Non-Technical Users

This study explores the potential of LangChain, a framework for constructing applications with advanced language models, to translate natural language queries into executable SQL code. Study propose an innovative LangChain-based architecture that receives a natural language query, analyzes it with a language model, and generates the corresponding SQL statement for database querying. This approach aims to empower non-technical users, facilitate inter-team collaboration, and enable data-informed decision-making. However, challenges persist, including managing complex queries and grasping domain-specific terminology. This research investigates the methodology and system design of our proposed natural language interface for databases, leveraging LangChain and extensive language models. Study also explore the possibilities and potential applications of this system, as well as future research avenues for enhancing its functionalities and addressing current constraints. By integrating advanced natural language processing with database technologies, this research aims to enable inclusive and powerful data querying experiences.

Engaging evaluation champions: strategies to build evaluation capacity and promote data-informed decision making in youth development

Youth development researchers and practitioners share an interest in ensuring that youth development programs contribute positively to youth outcomes. Engaging in evaluation and data-informed decision making (DIDM) has the potential to empower practitioners to improve and adapt programs, improving youth outcomes. Yet, not all practitioners are comfortable engaging in evaluation and/or DIDM. Engaging in partnerships and utilizing internal supporters of evaluations (or “champions”) have been identified as potential strategies to build evaluation capacity and strengthen DIDM within an organization. However, little research explores how to do so. This study engaged evaluation champions in four states to examine their experience as they partnered with practitioners within their organization. Results suggest that peer groups can be utilized to promote evaluation capacity, especially by utilizing peers that already have an interest in evaluation. Practitioner engagement can also be developed by using less academic jargon in communication, highlighting the practical value of evaluation, and building capacity slowly.

Meritus Health: Driving Data-Informed Decision-Making at Every Level.

With so much data available, health system leaders are challenged with sifting through it all to find the most useful information for decision-making. Meritus Health implemented effective approaches to understand, use, and communicate large amounts of data to alleviate some of this burden. These processes include system-wide daily huddles, dashboards, and standardized communication write-ups.

Using Text Mining to Illuminate How Community and Technical Colleges Adapt During COVID-19

ABSTRACT The COVID-19 crisis has caused unprecedented disruptions to all facets of postsecondary education. As open-access institutions serving diverse students, community and technical colleges have responded to pandemic-imposed challenges through numerous adaptations. To bring clarity and purpose to community and technical colleges’ institutional decision-making during times of crisis, we used text mining techniques to identify patterns underlying change and adaptations since March 2020 based on text data from weekly newsletters published by a 2-year college system in a Midwestern state. Our findings indicate that community and technical colleges strove to meet their core institutional mission and centered student needs during the crisis. Specifically, the colleges emphasized accessibility, flexibility, and continuity of education. However, we also identified a lack of depiction of important issues, such as support for faculty/staff and an explicit focus on issues of diversity, equity, and inclusion. Our study also demonstrates text mining techniques as an innovative toolkit to support community and technical colleges with data-informed decision-making to enhance institutional effectiveness.

P3LS: Partial Least Squares under privacy preservation

Modern manufacturing value chains require intelligent orchestration of processes across company borders in order to maximize profits while fostering social and environmental sustainability. However, the implementation of integrated, systems-level approaches for data-informed decision-making along value chains is currently hampered by privacy concerns associated with cross-organizational data exchange and integration. We here propose Privacy-Preserving Partial Least Squares (P3LS) regression, a novel federated learning technique that enables cross-organizational data integration and process modeling with privacy guarantees. P3LS involves a singular value decomposition (SVD) based PLS algorithm and employs removable, random masks generated by a trusted authority in order to protect the privacy of the data contributed by each data holder. We demonstrate the capability of P3LS to vertically integrate process data along a hypothetical value chain consisting of three parties and to improve the prediction performance on several process-related key performance indicators. Furthermore, we show the numerical equivalence of P3LS and PLS model components on both a synthetic and a real-world dataset and provide a thorough privacy analysis of the former. Moreover, we propose privacy-preserving explained X- and Y-block variance computations for determining the contribution of each data holder to the federated process model as a basis to incentivize data federation and fair profit-sharing.

Interpretable Bike-Sharing Activity Prediction with a Temporal Fusion Transformer to Unveil Influential Factors: A Case Study in Hamburg, Germany

Bike-sharing systems (BSS) have emerged as an increasingly important form of transportation in smart cities, playing a pivotal role in the evolving landscape of urban mobility. As cities worldwide strive to promote sustainable and efficient transportation options, BSS offer a flexible, eco-friendly alternative that complements traditional public transport systems. These systems, however, are complex and influenced by a myriad of endogenous and exogenous factors. This complexity poses challenges in predicting BSS activity and optimizing its usage and effectiveness. This study delves into the dynamics of the BSS in Hamburg, Germany, focusing on system stability and activity prediction. We propose an interpretable attention-based Temporal Fusion Transformer (TFT) model and compare its performance with the state-of-the-art Long Short-Term Memory (LSTM) model. The proposed TFT model outperforms the LSTM model with a 36.8% improvement in RMSE and overcomes current black-box models via interpretability. Via detailed analysis, key factors influencing bike-sharing activity, especially in terms of temporal and spatial contexts, are identified, examined, and evaluated. Based on the results, we propose interventions and a deployed TFT model that can improve the effectiveness of BSS. This research contributes to the evolving field of sustainable urban mobility via data analysis for data-informed decision-making.

Decision Making in the Innovation Process: Data-Driven vs. Data-Informed

There has been a growing trend in the use of data-related buzzwords, and “data-driven decision-making” is one of them. This buzzword is often confused with “data-informed decision-making,” emphasizing the need to understand the role of data for effective decision-making. The article explains this misconception through tables and insights from experts like Geoffrey Moore, Tendayi Viki and Alexander Osterwalder. It emphasizes the need for a balanced approach, using both qualitative and quantitative data, and suggests starting with qualitative insights before moving to quantitative analysis. Ultimately, it stresses the importance of aligning organizational structures to leverage data effectively for innovation.

Use of responsible artificial intelligence to predict health insurance claims in the USA using machine learning algorithms

Aim: This study investigates the potential of artificial intelligence (AI) in revolutionizing healthcare insurance claim processing in the USA. It aims to determine the most effective machine learning (ML) model for predicting health insurance claims, leading to cost savings for insurance companies. Methods: Six ML algorithms were used to predict health insurance claims, and their performance was evaluated using various metrics. The algorithms examined include support vector machine (SVM), decision tree (DT), random forest (RF), linear regression (LR), extreme gradient boosting (XGBoost), and k-nearest neighbors (KNN). The research involves a performance assessment that encompasses key metrics. Additionally, a feature importance analysis is conducted to illuminate the critical variables that exert influence on the prediction of insurance claims. Results: The findings demonstrate that the XGBoost and RF models outperformed the other algorithms, displaying the highest R-squared values of 79% and 77% and the lowest prediction errors. The feature importance analysis underscores the pivotal role of variables such as smoking habits, body mass index (BMI), and blood pressure levels in the domain of insurance claim prediction. These results emphasize the degree to which these variables should be included in the formulation of insurance policies and pricing strategies. Conclusions: This study supports the transformative potential of AI, with specific emphasis on the XGBoost model, in extending the precision and efficiency of healthcare insurance claim processing. The identification of key variables and the mitigation of prediction errors not only signal the potential for substantial cost savings but also affirm the potential to integrate AI into healthcare insurance processes. This research supports the value of the utilization of AI as an emerging tool for process optimization and data-informed decision-making within the healthcare insurance domain.

Effect of Monitoring and Evaluation (M&E) Practices on Performance of Health Funded Projects in Rwanda: A Case of Malaria Control, Maternal & Child Health Programs of Ministry of Health

The study assessed the effect of Monitoring and Evaluation (M&E) practices on performance of health funded projects in Rwanda: a case of Malaria control, Maternal & Child Health programs in Ministry of Health. It had three specific objectives: to assess the effect of M&E planning on project performance, to examine the influence of M&E data collection and analysis, and to evaluate the impact of M&E reporting and decision making. The hypotheses associated with each of these objectives were supported by the research findings, indicating significant effects in each area. This study used a descriptive survey and correlative research designs, and also the mixed approaches including qualitative and quantitative approaches. The population was 137 people working with Malaria control, Maternal & Child Health programs in Ministry of Health in Rwanda. The study used stratified and universal sampling techniques to select all 137 respondents as sample size. Data-collection instruments were testing questionnaires, interview schedules or guides, rating scales, and survey plans or any other forms which were used to collect information on substantially identical items or more respondents. The observation was done on situation success of health project management and M&E practices. Documentary technique was used by the researcher to obtain secondary information such as reports of Malaria control, Maternal & Child Health programs in Ministry of Health in Rwanda. The models showed a strong positive relationship between M&E planning, data collection and analysis, reporting, and decision making and project performance. This relationship was supported by a high correlation coefficient (R), with approximately 83.2% of the variance in project performance explained by these predictors. However, the presence of potential autocorrelation in the residuals suggests that further investigation is needed to ensure the validity of the model. In conclusion, the study effectively supported all three research hypotheses, confirming the significant impact of M&E planning, data collection and analysis, and reporting and decision making on the performance of Malaria control and Maternal & Child Health programs within the Ministry of Health. The Ministry of Health should invest in comprehensive M&E planning, setting clear objectives, defining indicators, and establishing a solid framework for improved performance. Timely and informative reporting of M&E results is crucial. The Ministry should prioritize the communication of results to facilitate data-informed decision-making.

Efficacy and Safety of Ponesimod Compared with Teriflunomide in Female Patients with Relapsing Multiple Sclerosis: Findings from the Pivotal OPTIMUM Study.

Background: Multiple sclerosis (MS) is threefold more prevalent in women than men. However, sex-specific efficacy analysis for MS disease-modifying therapies is not typically performed. Methods: Post hoc analyses of data from female patients enrolled in the phase 3, double-blind OPTIMUM study of relapsing MS were carried out. Eligible adults were randomized to ponesimod 20 mg or teriflunomide 14 mg once daily for up to 108 weeks. The primary endpoint was annualized relapse rate (ARR); secondary endpoints included change in symptom domain of Fatigue Symptom and Impact Questionnaire-Relapsing Multiple Sclerosis (FSIQ-RMS) at week 108, number of combined unique active lesions (CUALs) per year on magnetic resonance imaging, and time to 12- and 24-week confirmed disability accumulation (CDA). Results: A total of 735 female patients (581 of childbearing potential) were randomized to ponesimod (n = 363, 49.4%) or teriflunomide (n = 372, 50.6%). Relative risk reduction in the ARR for ponesimod versus teriflunomide was 33.1% (mean, 0.192 vs. 0.286, respectively; p < 0.002). Mean difference in FSIQ-RMS for ponesimod versus teriflunomide was -4.34 (0.12 vs. 4.46; p = 0.002); rate ratio in CUALs per year, 0.601 (1.45 vs. 2.41; p < 0.0001), and hazard ratio for time to 12- and 24-week CDA risk estimates, 0.83 (10.7% vs. 12.9%; p = 0.38) and 0.91 (8.8% vs. 9.7%; p = 0.69), respectively. Incidence of treatment-emergent adverse events was similar between treatment groups (89.0% and 90.1%). Conclusions: Analyses demonstrate the efficacy and safety of ponesimod, versus active comparator, for women with relapsing MS, supporting data-informed decision-making for women with MS. Clinical Trial Registration Number: NCT02425644.

Improving the Quality and Use of Malaria Surveillance Data: Results from Evaluating an Integrated Malaria Information Storage System at the Health Facility Level in Selected Districts in Mozambique.

Mozambique addressed critical malaria surveillance system challenges by rolling out an integrated malaria information storage system (iMISS) at the district level in February 2021. The iMISS integrates malaria data from existing systems across thematic program areas to improve data availability and use. In seven districts, the platform was extended to health facilities (HFs), allowing HFs to access iMISS and use tablets to submit monthly malaria reports to a central database, eliminating the need for paper-based reporting to districts. A structured evaluation of the iMISS rollout to HFs was carried out in February-July 2021. The four evaluation areas were data quality (reporting rate, timeliness, and fidelity) of monthly malaria reports electronically submitted to the iMISS, adoption of the iMISS for data-informed decision-making, system maintenance, and acceptability of the iMISS among target users. All 94 HFs in the seven targeted districts were assessed. Over the 6-month period, 86.1% of reported cases on the iMISS were consistent with cases recorded in paper-based reports, allowing for up to 10% discrepancy. In addition, 69.0% of expected monthly district meetings were held, and information from iMISS was discussed during 58.6% of these meetings. Maintenance issues, mostly related to tablet access and internet connectivity, were experienced by 74.5% of HFs; 33.7% of issues were resolved within 1 month. The iMISS and electronic submission of malaria reports were well accepted by HF- and district-level users. Continued political commitment and timely execution of issue management workflows are crucial to ensure trust in the new platform and facilitate higher levels of data use.

College Major Area and Career Commitment: Rethinking STEAM in Educational-Vocational Guidance

Objective: The exigency for higher education to exhibit outcomes aligning with career competencies has intensified, driven by external pressures favoring job-specific training. Amidst shifting career tendencies of Generations Y and Z and the advent of artificial intelligence-led automation, the valuation of different college majors has come under scrutiny. This study aims to dissect the prevailing assumptions and explore the satisfaction and career commitment levels among individuals with career-focused degrees. Methods: Utilizing a mixed-methods approach, the study delves into the relationships among areas of study, career commitment, self-esteem, and self-efficacy among alumni from a private Midwestern liberal arts college. Instruments like Core Self-Evaluations, Career Commitment, and Career Reconsiderations facilitated data collection from graduates over the past three decades. Results: Findings underscore higher career connectedness for Arts and Humanities and Physical and Social Sciences graduates compared to Business majors. Conversely, Business majors exhibited a significant surge in career reconsideration, suggesting a potential disconnect or regret among this cohort, possibly owing to the dynamic nature of the business field. Conclusion: The data advocates for a recalibrated approach in vocational guidance aligning with educational majors, promoting data-informed decision-making among stakeholders. The variation in career connectedness and reconsideration across different study fields necessitates a nuanced approach in educational and vocational counseling to enhance alignment with long-term career satisfaction and commitment, preparing students aptly for the evolving job market landscape.