Data-driven Insight Research Articles

Learning, Integration, and Re-Configuration: Dynamic Capabilities as Drivers of Data-Driven Insights and Decision Quality

Purpose The study investigates the role of the dynamic capabilities of learning, integration, and re-configuration on the quality of decision-making mediated by data-driven insights. It particularly examines how these dynamic capabilities individually contribute to offering efficient and effective decision-making in the hospitality sector. Design/methodology/approach This study employs a quantitative research approach harnessed through surveys and the application of partial least square structural equation modelling using SMART PLS 4. The hospitality sector being a highly competitive industry was selected for data collection. Findings The findings reveal that learning and reconfiguration capabilities have a significant positive individual impact on generating data-driven insights hence improving the effectiveness and efficiency of decision-making. Contrarily, the integration capabilities standalone did not influence in pointing out the statistical significance of the relationship with data-driven insight or decision-making quality, an indication that integration has to be combined with other dynamic capabilities for it to be effective. Originality/Value This study adds novelty to the literature by examining the individual influences of dynamic capabilities—learning, integration, and reconfiguration on data-driven insights and decision-making quality. Unlike previous studies that have generally explored bundles of capabilities, this study isolates each to answer the more nuanced question of how organizations can, in a strategic manner, invest in certain dynamic capabilities that can improve decision-making.

Three Faces of Heroism: An Empirical Study of Indirect Literary Translation Between Chinese-English-Portuguese of Wuxia Fiction

Abstract This paper conducts an empirical study using digital humanities to explore the indirect translation of Chinese contemporary literature, focusing on a Portuguese version of Jin Yong’s A Hero Born: The Legend of the Condor Heroes derived from its English translation. Employing BERTopic modeling, it quantifies thematic changes related to heroism in the translation process from Chinese through English to Portuguese. This study examines how Chinese concepts of heroism are adapted or rewritten in translation, exploring the notion of intermediate translations as “secondhand” and “distorted” from digital humanities perspective. By applying the zero-shot BERTopic model, the research allows for a quantitative comparative analysis of topic distribution across the original, intermediate, and final translation texts. The findings aim to enrich translation studies by demonstrating how Chinese cultural concepts are transformed and circulated through indirect translation, providing a data-driven insight into the global dissemination and reception of Wuxia literature.

Data-Driven Management: The Impact of Big Data Analytics on Organizational Performance

Inevitably, in such a fast-moving landscape of big data analytics lies the transformative opportunity for any organization to unlock new avenues of growth, operational efficiency, and strategic decision-making. This paper contributes a comprehensive methodology that will help businesses take advantage of big data analytics to secure a continued competitive advantage. At the core of this methodology is to have a robust data governance framework that will establish the security, integrity, and accessibility of the enterprise data assets by specifying relevant policies, processes, and technologies. This, in turn, within such a framework, allows state-of-the-art AI-driven anomaly detection mechanisms for encryption and access control to implement protection measures around sensitive information while enabling secure yet efficient data utilization. The methodology also continues its approach of putting in place an integrated data ecosystem that brings together different pockets or sources of data, such as real-time operation data, customer interaction, and unstructured information. A critical component of this methodology is putting in place the advanced predictive analytics capabilities that could be run based on tapping into the power of machine learning algorithms; by doing so, the organization will be predicting market trends, customer behavior, and risks highly accurately. This will be a very proactive way of making decisions that would let the business efficiently use the available resources, innovate products and services ahead of time, and create a distinctly competitive advantage. The transformative ability of this methodology for big data analytics opened new channels toward growth and innovation and firmly established the organization as an industry leader with long-lasting competitive advantage. The place of data-driven insight, data culture, and responsible data practice has been the key to success in this organization.

Digital Participatory Model as Part of a Data-Driven Decision Support System for Urban Vibrancy

Digital participation relies on computational systems as the instruments for expert engagement, data-driven insight, and informed decision-making. This study aims to increase expert engagement with the Bayesian-based decision support model in evaluating urban vibrancy decisions. In this study, urban vibrancy parameters are defined using “economic, use, and image value” measures. This article focuses on the visual aspect of vibrancy, defined as the image value of place. The image value is evaluated through likability and likability features. The case study area is the Eminönü Central Business District in the Istanbul Historic Peninsula due to its distinctive urban dynamics derived from the duality of being a cultural and cosmopolitan city center. This research presents a method as a decision support system (DSS) model based on the Bayesian belief network (BBN) and spatial BBN for supporting urban vibrancy decisions. The spatial BBNs monitor spatial outcomes of variables’ dependencies that form through the BBN relationship network. Spatial BBN tools monitors the spatial impact of decisions for informed urban interventions. The results demonstrate that urban greening, pedestrianization, and human-scaled streetscapes should be prioritized to make streets more likable. The most significant intervention areas are Tahtakale for signboard regulation, Sultanahmet and Vefa for cultural landscape improvement, and Vefa and Mahmutpaşa for planning building enclosures. The participation is achieved by evaluating urban vibrancy with what-if scenarios using BBN. The developed DSS model addresses which parameters should be prioritized, and what are their spatial consequences. The use of spatial BBN tools presents certain limitations in terms of interoperability and user interaction. Overall, this research contributes to participatory urban planning by incorporating both conditional and spatial dependencies. This unique approach not only promotes a more holistic understanding of urban vibrancy but also contributes to the advancement of digital participation in urban planning decisions.

The Importance of Consumer Insights for Precision Marketing in the Era of Big Data

E-commerce developed so far has gradually changed from a crude incremental dividend to a refined operation, which examines the operation, marketing, technology, and serviceability of enterprises due to the highly intense competition. To maintain an advantage in the competition and achieve business growth, enterprises need to improve their influence continuously and seize opportunities to attract consumers attention. Precision marketing has become one of the critical factors. The prerequisite for precision marketing is to have precise insights into consumers. The arrival of the big data era provides new ways and means for enterprises to gain insight into consumers. This article mainly emphasizes that enterprises can make use of all kinds of data, from collecting and organizing consumer data to final formulating marketing strategies and improving the marketing strategy through consumer insight research, rather than using the traditional method of depicting consumer profiles and one-sided formulation of marketing strategy. The article introduces the big data era, explains the necessity of consumer insight, and states the relationship between big data and precision marketing. Additionally, it describes consumer insight, mainly from utilizing big data to analyze the consumers and understand the advantages and potential loopholes of big data-driven consumer insight. With the popularization of the Internet and mobile devices, the traces left by users are becoming more comprehensive and wealthier. It is crucial for enterprises to conduct data analysis to dig and analyze consumer habits, experiences, and values to tightly link consumer insights and marketing communications to carry out precision marketing.

Machine learning-enabled performance prediction and optimization for iron-chromium redox flow batteries.

Iron-chromium flow batteries (ICRFBs) are regarded as one of the most promising large-scale energy storage devices with broad application prospects in recent years. However, transitioning from laboratory-scale development to industrial-scale deployment can be a time-consuming process due to the multitude of complex factors that impact ICRFB stack performance. Herein, a data-driven optimization methodology applying active learning, informed by an extensive survey of the literature encompassing diverse experimental conditions, is proposed to enable exceptional precision in predicting ICRFB system performance considering both operation conditions and key materials selection. Specifically, multitask ML models are trained on experimental data with a high prediction accuracy (R2 > 0.92) to link ICRFB properties to energy efficiency, coulombic efficiency, and capacity. We also interpret the ML models based on Shapley additive explanations and extract valuable insights into the importance of descriptors. It is noted that the operation conditions (current density and cycle number) and the electrode type are the most critical descriptors affecting the voltage efficiency and coulombic efficiency while the electrode size strongly affects the capacity. Moreover, active learning is used to explore the most optimized cases considering the highest energy efficiency and capacity. The versatility and robustness of the approach are demonstrated by the successful validation between ML prediction and our experiments of energy efficiency (±0.15%) and capacity (±0.8%). This work not only affords fruitful data-driven insight into the property-performance relationship, but also unveils the explainability of critical properties on the performance of ICRFBs, which accelerates the rational design of next-generation ICRFBs.

The use of data-driven insight in ambidextrous digital transformation: How do resource orchestration, organizational strategic decision-making, and organizational agility matter?

Drawing on the literature about digital transformation and ambidexterity, our study conceptualizes ambidextrous digital transformation and investigates the relevant antecedents. We apply and advance the business intelligence and analytics theory and the resource management theory in the context of digital transformation by integrating them together. In particular, we develop and verify a research model on how data-driven insight contributes to ambidextrous digital transformation. Based on data drawn from 312 firms undergoing digital transformation in China, our findings indicate distinct effects of data-driven insight exerted on efficiency transformation and value transformation with resource orchestration as a partial mediator, and the contingent role of diverse organizational strategic decision-making and organizational agility. Our study extends existing literature by conceptualizing and developing a two-component construct of ambidextrous digital transformation and revealing how the critical organizational capabilities and behaviors underlying it work together. Our findings will stir up academic interest and curiosity to further unravel the ambidextrous path of digital transformation, and have managerial implications for managers and authorities committed to promote digital transformation and prosper digital economy.

Data-Driven Insight into the Reductive Stability of Ion-Solvent Complexes in Lithium Battery Electrolytes.

Lithium (Li) metal batteries (LMBs) are regarded as one of the most promising energy storage systems due to their ultrahigh theoretical energy density. However, the high reactivity of the Li anodes leads to the decomposition of the electrolytes, presenting a huge impediment to the practical application of LMBs. The routine trial-and-error methods are inefficient in designing highly stable solvent molecules for the Li metal anode. Herein, a data-driven approach is proposed to probe the origin of the reductive stability of solvents and accelerate the molecular design for advanced electrolytes. A large database of potential solvent molecules is first constructed using a graph theory-based algorithm and then comprehensively investigated by both first-principles calculations and machine learning (ML) methods. The reductive stability of 99% of the electrolytes decreases under the dominance of ion-solvent complexes, according to the analysis of the lowest unoccupied molecular orbital (LUMO). The LUMO energy level is related to the binding energy, bond length, and orbital ratio factors. An interpretable ML method based on Shapley additive explanations identifies the dipole moment and molecular radius as the most critical descriptors affecting the reductive stability of coordinated solvents. This work not only affords fruitful data-driven insight into the ion-solvent chemistry but also unveils the critical molecular descriptors in regulating the solvent's reductive stability, which accelerates the rational design of advanced electrolyte molecules for next-generation Li batteries.

Data-Driven Insights on Time-to-Failure of Electromechanical Manufacturing Devices: A Procedure and Case Study

Nowadays, there is a fresh push towards putting more attention on sustainability issues without affecting productivity as main target in Industrial Cyber-Physical Systems (ICPS). In this direction, this work proposes a procedure and presents a data-driven insight method in order to predict the remaining useful life (RUL) and to classify faults by a condition base-monitoring (CbM). Therefore, by using a framework that combines both outputs, a maintenance stop can be scheduled near to the failure, thus improving its sustainability, without affecting productivity. A fuzzy decision-making strategy supported on generated insights is developed in order to extend the useful life of electromechanical devices. A case study is presented in order to assess the proposed methodology using a dataset of bearing faults. Experimental results and its comparison with previous reported works corroborate a good trade-off solution offered by the proposed procedure considering productivity and sustainability for bearing faults detection.

Synergistic Effect of As(III)/Fe(II) Oxidation by Acidianus brierleyi and the Exopolysaccharide Matrix for As(V) Removal and Bioscorodite Crystallization: A Data-Driven Modeling Insight

Bioscorodite crystallization is a promising process for the proper immobilization of arsenic from acidic metallurgical wastewater, and Acidianus brierleyi is an effective archaeon to oxidize Fe(II) and As(III) simultaneously. This paper deals with the development of an experimentally validated mathematical model to gain insight into the simultaneous processes of Fe(II) and As(III) oxidation via microbial cells and the exopolysaccharide (EPS) matrix, As(V) precipitation, and bioscorodite crystallization, which are affected by several factors. After the mathematical structure was proposed, a model fitting was performed, finding global determination coefficients between 0.96 and 0.99 (with p-values < 0.001) for all the variables. The global sensitivity analysis via Monte Carlo simulations allowed us to identify the critical parameters whose sensitivity depends on culture conditions. The model was then implemented to evaluate the effect of cell concentration, Fe(II) and As(III) concentrations (at Fe/As = 1.4), and oxidation rate constants for A. brierleyi and the EPS region, noting that these factors play an important role in the process. Our results showed that the proposed model can be used as a robust simulation platform for the further analysis of the bioscorodite crystallization process under extremophilic conditions.

Biomedical evidence engineering for data-driven discovery.

With the rapid development of precision medicine, a large amount of health data (such as electronic health records, gene sequencing, medical images, etc.) has been produced. It encourages more and more interest in data-driven insight discovery from these data. A reasonable way to verify the derived insights is by checking evidence from biomedical literature. However, manual verification is inefficient and not scalable. Therefore, an intelligent technique is necessary to solve this problem. This article introduces a framework for biomedical evidence engineering, addressing this problem more effectively. The framework consists of a biomedical literature retrieval module and an evidence extraction module. The retrieval module ensembles several methods and achieves state-of-the-art performance in biomedical literature retrieval. A BERT-based evidence extraction model is proposed to extract evidence from literature in response to queries. Moreover, we create a dataset with 1 million examples of biomedical evidence, 10000 of which are manually annotated. Datasets are available at https://github.com/SendongZhao.

Transient segregation of ternary granular mixtures simultaneously charged into and discharged from a shaft furnace

The simultaneous charging at the top and discharging at the bottom inevitably have the combined effects on granular segregation in the shaft furnace. Herein, we first establish a 1:1 full mathematical model of shaft furnace and validate the parameters for the discrete element method (DEM) simulation. Then a quantified method based on an equilateral triangle is updated to characterize the ternary-component (coke, sinter, and pellet) granular segregation. The results show that compared with the sole operation, the simultaneous charging and discharging operations make the overall ternary segregation worse in the central and peripheral regions of the shaft furnace. Furthermore, transient granular behaviors in terms of the granular profile, the instantaneous segregation, the particle-scale trajectory and vertical velocity, and the granular number distribution are given full discussions to provide a data-driven insight into the combined effects on granular segregation.

MDLF: A Multi-View-Based Deep Learning Framework for Individual Trip Destination Prediction in Public Transportation Systems

Understanding and predicting each individual’s real-time travel destination given the origin information in urban public transportation systems is crucial for personalized traveler recommendation, targeted demand management, dynamic traffic operations and so on. Existing methods are often based on modeling the regular travel patterns through analyzing the long-term personal travel information. They are suitable for destination prediction of individual regular trips with regular travel patterns, but may not work well for occasional trips with strong randomness and uncertainty, especially for the individuals with a few historical travel data. In this paper, we focus on more challenging issue about destination prediction of occasional trips. We design a general Multi-View Deep Learning Framework (MDLF) based on the data-driven insight that a location where a user will destine to is not only related to the user’s own travel preference to the location, but also influenced by crowd’s travel preference and the region’s characteristics of the location under certain spatiotemporal contexts. The destination of an individual’s occasional trip can be predicted by combining all these complementary influencing factors. The novelty of MDLF is mainly reflected in two aspects. The first is the effective feature extraction from multiple and complementary views. The second is that a CNN (Recurrent Neural Network) based deep learning component for predicting each occasional trip’s destination by calculating a moving trend score for each possible destination. We evaluate the MDLF based on two real-world smart card datasets collected by AFC (Automatic Fare Collection) Systems. The experimental results demonstrate the superiority of MDLF against other competitors.

A Systematic Review of Federated Learning in the Healthcare Area: From the Perspective of Data Properties and Applications

Recent advances in deep learning have shown many successful stories in smart healthcare applications with data-driven insight into improving clinical institutions’ quality of care. Excellent deep learning models are heavily data-driven. The more data trained, the more robust and more generalizable the performance of the deep learning model. However, pooling the medical data into centralized storage to train a robust deep learning model faces privacy, ownership, and strict regulation challenges. Federated learning resolves the previous challenges with a shared global deep learning model using a central aggregator server. At the same time, patient data remain with the local party, maintaining data anonymity and security. In this study, first, we provide a comprehensive, up-to-date review of research employing federated learning in healthcare applications. Second, we evaluate a set of recent challenges from a data-centric perspective in federated learning, such as data partitioning characteristics, data distributions, data protection mechanisms, and benchmark datasets. Finally, we point out several potential challenges and future research directions in healthcare applications.

Data-driven insight into the puzzle-based cybersecurity training

Puzzle-based training is a common type of hands-on activity accompanying formal and informal cybersecurity education, much like programming or other IT skills. However, there is a lack of tools to help the educators with the post-training data analysis.Through a visualization design study, we designed the Training Analysis Tool that supports learning analysis of a single hands-on session. It allows an in-depth trainee comparison and enables the identification of flaws in puzzle assignments. We also performed a qualitative evaluation with cybersecurity experts and students. The participants apprised the positive influence of the tool on their workflows. Our insights and recommendations could aid the design of future tools supporting educators, even beyond cyber security.

Measuring reopening readiness: a universal COVID-19 index for US states

PurposeWhile United States is among countries with the world’s highest coronavirus infections, its approaches and policies to reopen the economy vary by state. A lack of objective criteria and monitoring toward satisfying the criteria can lead to another COVID-19 outbreak and business closures. Considering the pressing need to return to normalcy without a rebound of COVID-19 infections and deaths, an index that provides a data-driven and objective insight is urgently needed. Hence, a method was devised to assess the severity of the COVID-19 pandemic and determine the degree of progress any state has made in containing the spread of COVID-19.Design/methodology/approachUsing measures such as the weekly averages of daily new deaths, ICU bed occupancy rates, positive cases and test positivity rates, two indexes were developed: COVID-19 reopening readiness and severity.FindingsA clear difference in the pandemic severity trends can be observed between states, which is possibly due to the disparity in the state’s response to coronavirus. A sharp upward trend in index values requires caution prior to moving to the next phase of reopening.Originality/valueThe composite indexes advanced in this study will provide a universal, standardized and unbiased view of each state’s readiness to reopen and allow comparisons between states. This in turn can help governments and health-care agencies take counter measures if needed as to the anticipated demand for future health-care services and minimize adverse consequences of opening.

Leveraging simulated and empirical data-driven insight to supervised-learning for porosity prediction in laser metal deposition

The advent of digital-twin manufacturing in additive manufacturing (AM) is to integrate the physical world of real-time 3D printing with the digital world of a simulated print. This paper contributes to digital-twin manufacturing in laser-based additive manufacturing by combining melt pools’ simulated thermal behavior via finite element analysis (FEA) and melt pools’ empirical thermal behavior via pyrometry-based sensors. Studying the thermal behavior of melt pools based on heat transfer characteristics determines melt pools’ porosity and part quality. FEA uses Godak’s moving heat flux to capture the melt pools’ physically bound temperature profile in three dimensions. Simulated data helps to mitigate the influence of measuring errors from real-world data and provides non-observable data such as gradient changes of thermal behavior at the curvature of the 3D melt pool. The pyrometer captures empirical temperature behavior, including uncertainty and randomness introduced to the process. A significant knowledge gap exists when predicting melt pool porosity accurately with theoretical FEA and empirical in situ evidence alone. The gap is bridged by combining the data sources, specifically, feature engineering via functional principal component analysis (empirical data source) and capturing the melt pool's 3-D temperature shape profile via FEA (simulated data source). A hybrid model predicts melt pool porosity by capturing the strengths of prior simulated and posterior in situ empirical data by matching simulated melt pools to real-world empirical melt pools. Moreover, comparing predicted porosity labels with true porosity labels of Ti–6Al–4V thin-wall structure from laser metal deposition verified the proposed interpretable and robust supervised-learning model's validity. This methodology can apply to other materials and part shapes printed under various additive-manufactured printers.

Data Storytelling with Dashboard: Accelerating Understanding Through Data Visualization in Financial Technology Company Case Study

In the rapid development many organisation rely on context data to support as well as to assist its decision making process. Consequently, Business Intelligence (BI), Dashboard, and Data Visualization emerged as primary tools in early 1990s as a way to help practitioners, data analyst, and data scientist to present context data into an actionable information for decision making process. However, despite its robust and powerful tools, recent study done by Kaggle’s survey in 2017 resulted that in the last five years, many companies were not able to create effective data-driven dashboard due to complex dataset, poor dashboard design, and insufficient storytelling. Hence, understanding of who is going to use dashboard, choosing which data and metrics to visualize in the right context, knowing how to convey information, driving engagement, and persuading audiences are essential in current business practices. This study is aimed to help practitioners to understand the impact of effective dashboard can have on decision making process, to design leveraging dashboard, and to present the dashboard in storytelling. A literature study is performed to gather all relevant information resulted in guidelines for dashboard creator. Case study in financial technology company is applied to experiment and to test the guidelines for assisting dashboard creator to present data-driven insight to the stakeholder.

‘Risk’ in political discourse. A corpus approach to semantic change in German Bundestag debates

Our paper presents a diachronic corpus linguistic approach to the conceptualisation and framing of ‘risk’ in German parliamentary discourse. We analyse all occurrences of risk based on the complete collection of plenary protocols of the German Bundestag covering the legislative periods 1–18 (1949–2017). We apply methods used in Digital Discourse Analysis to show how the concept of ‘risk’ changes through time. Therefore, we investigate co-occurrences to find typical collocations and contexts. We measured three peaks: the first in the late 1950s, the second from the 1970s onwards, and the third in the first decade of the new millennium. We show that ‘risk’ as a negative and ‘chance’ as a positive concept have partly taken the communicative place of ‘danger’ and ‘possibility’. The collocates of risk that can be related to different knowledge domains increase until 2009. This hints at the growing importance of risk thematizations in more and more thematic contexts. Technology and economics turn out to be the most important of these contexts. The analysis shows a clear tendency towards a negative and rather generic use of the word risk. Our results provide for the first time a data-driven insight into the long-term development of the ‘risk’ concept in political discourse of the Federal Republic of Germany.

Big data analytics capability and decision-making: The role of data-driven insight on circular economy performance

Big data analytics (BDA) is a revolutionary approach for sound decision-making in organizations that can lead to remarkable changes in transforming and supporting the circular economy (CE). However, extant literature on BDA capability has paid limited attention to understanding the enabling role of data-driven insights for supporting decision-making and, consequently, enhancing CE performance. We argue that firms drive decision-making quality through data-driven insights, business intelligence and analytics (BI&A), and BDA capability. In this study, we empirically investigated the association of BDA capability with CE performance and examined the mediating role of data-driven insights in the relationship between BDA capability and decision-making. Data were collected from 109 Czech manufacturing firms, and partial least squares structural equation modeling was applied to analyze the data. The results reveal that BDA capability and BI&A are positively associated with decision-making quality. This effect is stronger when the manufacturer utilizes data-driven insights. The results demonstrate that BDA capability drives decision-making quality in organizations, and data-driven insights do not mediate this relationship. BI&A is associated with decision-making quality through data-driven insights. These findings offer important insights to managers, as they can act as a reference point for developing data-driven insights with the CE paradigm in organizations.