Data-driven Exploration Research Articles

Enhancing Classroom Engagement: A Data-Driven Exploration of EdTech Integration Using Machine Learning

Enhancing Classroom Engagement: A Data-Driven Exploration of EdTech Integration Using Machine Learning

An Overview and Future Outlook of the Murder Mystery Game Industry in China

Murder mystery games have experienced a notable surge in popularity, with their origins deeply rooted in storytelling traditions. This research paper delves into the reasons for their rise, emphasizing the influences of macroeconomic factors, educational trends, and digitalization. These elements not only provided a backdrop for the murder mystery game's resurgence but also played a pivotal role in shaping its contemporary form. A comparative analysis of online and offline modalities reveals distinct user experiences and commercial potentials for each. Furthermore, the research offers pragmatic suggestions for entrepreneurs interested in establishing offline murder mystery stores, underscoring the nuances of market dynamics and consumer preferences. Harnessing statistical data from past survey results and well-informed estimations, the study presents a data-driven exploration into the genre's growth and transformation. By charting the game's historical trajectory and examining its present state, this paper provides a holistic overview of murder mystery games in the modern context and their significance in the evolving entertainment industry.

Personality–brain connection: Based on resting-state functional magnetic resonance imaging data-driven exploration

Background: The personality-brain association mechanism has been a topic of interest in the field of neuroscience. Usually, the previous research strategy was to first group the population based on different personality traits, and then explore the brain mechanisms corresponding to different personality groups. At present, a “brain-first” research strategy, which uses data-driven approaches instead of personality traits to first group the population, has been adopted to further enhance study objectivity. Methods: Here, we used a data-driven approach following the “brain-first” research strategy to deeply mine the resting-state brain functional magnetic resonance imaging data of 119 healthy participants, classified subjects into different groups based on brain image characteristics, and used the Sixteen Personality Factor Questionnaire to explain the variabilities of resting-state brain characteristics between different groups. Results: We have identified 3 personality–brain connections, including the privateness–left frontoparietal network, liveliness–sensory–motor network, and vigilance–sensory–motor network. Conclusion: We conclude that the above-mentioned three personality factors are based on brain neural activity, independent of the subjective experience of the personality scale creator, and have stronger explanatory power of brain imaging features.

Unlocking the Dynamics of Innovation Clusters Sectoral Impacts and Organisational Capabilities

Purpose: This article provides insights into how innovation clusters operate, how sectoral factors and organisational capabilities influence them, and how they contribute to economic growth and innovation in specific regions and industries. It provides insights into how sectoral effects, organisational capabilities, and collaboration dynamics intersect within these clusters, offering valuable information for policymakers and businesses seeking to leverage the potential of innovation clusters for regional and industry-specific development.
 Methodology/Approach: The research methodology involves a literature review, quantitative survey, factor analysis, cluster analysis, and correlation analysis to explore the relationships between sectoral effects, organisational capabilities, and collaboration within innovation clusters.
 Findings: The findings suggest that sector-specific effects influence cluster members' strategic and innovation capabilities. Technological complexity is predominant, impacting market orientation and product development capabilities. The article also identifies four distinct groups of organisations based on sectoral impacts and export activities within clusters.
 Research Limitation/implication: While the methodology provides valuable insights, there is potential sampling bias as non-random sampling via cluster manager recommendations may introduce selection bias. The findings are specific to Hungarian innovation clusters and may not be generalised universally.
 Originality/Value of paper: The uniqueness of this study lies in its holistic and data-driven exploration of innovation clusters, considering a wide range of factors and their practical implications for fostering collaboration and innovation in different industries.

Transition to Sustainable Energy System for Smart Cities—Literature Review

The article will contain a scientific analysis, showing thematic links between publications and consist of searching the data in the Scopus database. The timeframe of the searched publications will be 2010–2022. The parameters should also be narrowed down by selecting the following indexes: Science Citation Index Expanded (SCI-E), Social Science Citation Index (SSCI) and Emerging Sources Citation Index (ESCI), which was intended to eliminate abstracts of books and conference materials, leaving only reviewed works with the highest level of relevance for furthering knowledge. An additional limit of five minimum citations will be introduced. The analysis included 342 articles. Texts cited at least 100 times were highlighted. The research showed that authors focus piecemeal on selected aspects or problems, so an attempt was made to show thematic connections of keywords, illustrating the complexity of the transformations underway. The discussion identifies the most active authors and countries, especially exploring the topic of a transition to a sustainable energy system for smart cities. The authors have employed scientometric analysis to provide an objective and data-driven exploration of the transformation of sustainable energy systems for smart cities. This approach offers valuable insights into the research landscape, trends, and relationships within the field, which can guide future scientific research and contribute to a deeper understanding of the subject matter. As an additional element of this conclusion, the authors proposed supplementing the scientometric analysis with the foresight methodology. The authors’ research approach is distinguished by the following stages: problem formulation, data collection, choice of scientometric methodology, analysis of advantages and limitations of scientometrics, clustering analysis, data analysis, and presentation of results. Our systematic literature review systematizes the existing literature on the sustainable energy systems for smart cities, isolates main research interests, identifies future research avenues, and provides several important hints for researchers.

A Data-Driven Exploration of a New Islamic Fatwas Dataset for Arabic NLP Tasks

Islamic content is a broad and diverse domain that encompasses various sources, topics, and perspectives. However, there is a lack of comprehensive and reliable datasets that can facilitate conducting studies on Islamic content. In this paper, we present fatwaset, the first public Arabic dataset of Islamic fatwas. It contains Islamic fatwas that we collected from various trusted and authenticated sources in the Islamic fatwa domain, such as agencies, religious scholars, and websites. Fatwaset is a rich resource as it does not only contain fatwas but also includes a considerable set of their surrounding metadata. It can be used for many natural language processing (NLP) tasks, such as language modeling, question answering, author attribution, topic identification, text classification, and text summarization. It can also support other domains that are related to Islamic culture, such as philosophy and language art. We describe the methodology and criteria we used to select the content, as well as the challenges and limitations we faced. Additionally, we perform an Exploratory Data Analysis (EDA), which investigates the dataset from different perspectives. The results of the EDA reveal important information that greatly benefits researchers in this area.

Data-driven exploration of heterogeneous gasoline price elasticities using generalized random forests

Data-driven exploration of heterogeneous gasoline price elasticities using generalized random forests

Philanthropy in art: locality, donor retention, and prestige

A significant portion of funding for art comes from foundations, representing a key revenue stream for most art organizations. Little is known, however, about the quantitative patterns that govern art funding, limiting the fundraising efficiency of organizations in need of resources, as well as optimal funding allocation of donors. To address these shortcomings, here we relied on the IRS e-file dataset to identify $36B in grants from 46,643 foundations to 48,766 art recipients between 2010 and 2019, allowing us to quantify donor-recipient relationships in art. We find that philanthropic giving is broadly distributed, following a stable power-law distribution, indicating that some funders give considerably and predictably more than others. Giving is highly localized, with 60% of grants and funds going to recipients in the donor’s state. Furthermore, donors often support multiple local organizations that offer distinct artforms, rather than advancing a particular subarea within art. Donor retention is strong, with nearly 70% of relationships continuing the next year. Finally, we explored the role of institutional prestige in foundation giving, finding that funding does correlate with prestige, with notable exceptions. Our results present the largest and most comprehensive data-driven exploration of giving by foundations to art to date, unveiling multiple insights that could benefit both donors and recipients.

Daytime affect and sleep EEG activity: A data-driven exploration.

It has long been thought that links between affect and sleep are bidirectional. However, few studies have directly assessed the relationships between: (1) pre-sleep affect and sleep electroencephalogram (EEG) activity; and (2) sleep EEG activity and post-sleep affect. This study aims to systematically explore the correlations between pre-/post-sleep affect and EEG activity during sleep. In a community sample of adults (n = 51), we measured participants' positive and negative affect in the evening before sleep and in the next morning after sleep. Participants slept at their residence for 1 night of EEG recording. Using Fourier transforms, the EEG power at each channel was estimated during rapid eye movement sleep and non-rapid eye movement sleep for the full range of sleep EEG frequencies. We first present heatmaps of the raw correlations between pre-/post-sleep affect and EEG power during rapid eye movement and non-rapid eye movement sleep. We then thresholded the raw correlations with a medium effect size |r| ≥ 0.3. Using a cluster-based permutation test, we identified a significant cluster indicating a negative correlation between pre-sleep positive affect and EEG power in the alpha frequency range during rapid eye movement sleep. This result suggests that more positive affect during the daytime may be associated with less fragmented rapid eye movement sleep that night. Overall, our exploratory results lay the foundation for confirmatory research on the relationship between daytime affect and sleep EEG activity.

The nonlinear relationships between built environment features and urban street vitality: A data-driven exploration

The relationship between the built environment and urban street vitality, as a key issue of contemporary urban design, has been discussed over decades. However, most existing studies relying on linear regression models do not reveal the complicated impacts of built environment features and often neglect their threshold effects. As a response, this study applies the gradient boosting decision tree (GBDT) model with a large amount of new urban data to explore the in-depth understanding of urban street vitality. Based on the street samples from 12 Chinese cities, a series of morphological, functional, and human-scale features were analyzed together with socioeconomic indicators as control variables. The street vitality is measured by street activity intensity computed from billions of location-based service records. The results show that the nonlinear model brings an overall improvement in resolution. Specifically, compared with the functional and human-scale features, the morphological characteristics, especially the street intersection density, average block size, and building density, are dominant contributors to street vitality. It is also worth noting that most built-up environment features obtain the threshold effects on street vitality, which means there is a turning point where the effect of features changes. The interaction between built environment characteristic variables is common and can be divided into two typical types. Insights achieved in this study help to indicate an effective interval of built environment characteristics on vitality, which was missed in previous studies, and thus contribute to more precise urban design practices. Moreover, by clarifying the interaction influence mechanism, this study emphasizes the need for the planner to exploit synergies between variables through optimal combinations while avoiding their antagonistic effects.

Inverse-designed growth-based cellular metamaterials

Advancements in machine learning have sparked significant interest in designing mechanical metamaterials, i.e., materials that derive their properties from their inherent microstructure rather than just their constituent material. We propose a data-driven exploration of the design space of growth-based cellular metamaterials based on star-shaped distances. These two-dimensional metamaterials are based on periodically-repeating unit cells consisting of material and void patterns with non-trivial geometries. Machine learning models exploiting large datasets are then employed to inverse design growth-based metamaterials for tailored anisotropic stiffness. Firstly, a forward model is created to bypass the growth and homogenization process and accurately predict the mechanical properties given a finite set of design parameters. Secondly, an inverse model is used to invert the structure–property maps and enable the accurate prediction of designs for a given anisotropic stiffness query. We successfully demonstrate the frameworks’ generalization capabilities by inverse designing for stiffness properties chosen from outside the domain of the design space.

Extracting relevant predictive variables for COVID-19 severity prognosis: An exhaustive comparison of feature selection techniques.

With the COVID-19 pandemic having caused unprecedented numbers of infections and deaths, large research efforts have been undertaken to increase our understanding of the disease and the factors which determine diverse clinical evolutions. Here we focused on a fully data-driven exploration regarding which factors (clinical or otherwise) were most informative for SARS-CoV-2 pneumonia severity prediction via machine learning (ML). In particular, feature selection techniques (FS), designed to reduce the dimensionality of data, allowed us to characterize which of our variables were the most useful for ML prognosis. We conducted a multi-centre clinical study, enrolling n = 1548 patients hospitalized due to SARS-CoV-2 pneumonia: where 792, 238, and 598 patients experienced low, medium and high-severity evolutions, respectively. Up to 106 patient-specific clinical variables were collected at admission, although 14 of them had to be discarded for containing ⩾60% missing values. Alongside 7 socioeconomic attributes and 32 exposures to air pollution (chronic and acute), these became d = 148 features after variable encoding. We addressed this ordinal classification problem both as a ML classification and regression task. Two imputation techniques for missing data were explored, along with a total of 166 unique FS algorithm configurations: 46 filters, 100 wrappers and 20 embeddeds. Of these, 21 setups achieved satisfactory bootstrap stability (⩾0.70) with reasonable computation times: 16 filters, 2 wrappers, and 3 embeddeds. The subsets of features selected by each technique showed modest Jaccard similarities across them. However, they consistently pointed out the importance of certain explanatory variables. Namely: patient's C-reactive protein (CRP), pneumonia severity index (PSI), respiratory rate (RR) and oxygen levels -saturation Sp O2, quotients Sp O2/RR and arterial Sat O2/Fi O2-, the neutrophil-to-lymphocyte ratio (NLR) -to certain extent, also neutrophil and lymphocyte counts separately-, lactate dehydrogenase (LDH), and procalcitonin (PCT) levels in blood. A remarkable agreement has been found a posteriori between our strategy and independent clinical research works investigating risk factors for COVID-19 severity. Hence, these findings stress the suitability of this type of fully data-driven approaches for knowledge extraction, as a complementary to clinical perspectives.

High-throughput computational screening of adsorbents and membrane materials for acetylene capture

Adsorption-based and membrane-based separation technologies have attracted much attention in acetylene purification due to their low energy consumption and environmental friendliness. Herein, through high-throughput Monte Carlo simulations, we sweep through a total of 5039 materials available in crystal database and find that as many as 165 of them are potential acetylene-type adsorbents. Based on the first-principles force field, three structures with open metal sites were determined to have high adsorption capacity, surpassing that of CuBTC. An adsorption mode map of the trapped acetylene is depicted. Then, high-throughput molecular dynamics simulations were carried out to calculate the diffusion behavior of selected membrane materials and the Robeson Upper Bound for C2H2/CO2 was plotted. The diffusion path reveals the transition of the confined space of acetylene molecule from anion pillar to A-type channel and then to B-type channel in nanopores in a simple cubic topology. Moreover, data-driven exploration reveals that the H, O, and F atoms in the pores are preferential adsorption sites. As demonstration, the extreme gradient boosting model can accurately predict the loading of C2H2, and the decision tree model was used to show an optimal selection path (RE/T > 0.118, PLD ≤6.831 Å, GSA > 2724.7 m2/g) for the permeability of membrane materials.

Data-driven Exploratory Analysis for Raster Data Using Self-Organizing Maps Regressor

Sugarcane is one of the plantation commodities in Indonesia which has a big potential. Sugarcane growth consists of 4 phases that happen in a year. In the Grand Growth phase, sugarcane needs an appropriate condition to grow well and enter the next phase. The factors that affect sugarcane’s Grand Growth phase are water, temperature, and sunlight. Rainfall is one of the sugarcane water sources needed, but the rainfall intensity is different, and the rainfall distribution is uneven every year. The uneven rainfall caused water stress in a sugarcane plantation. That is why it is necessary to identify the water content in sugarcane plantations to maintain the quality of sugarcane. This study predicted the water content of sugarcane plantations so the areas indicated with water stress can be anticipated. Raster data are collected from Landsat-8 satellite imagery and analyzed using one of the data-driven exploration analysis methods, PCA (Principal Component Analysis), to analyze the overlay of the Landsat 8 imageries of the sugarcane plantation area. After that, the raster data were processed to calculate the water index of the sugarcane plantation, known as NDWI (Normalized Different Water Index). NDWI values of the sugarcane plantation area are converted into an array and then become data input for the Self-Organizing Map Regressor algorithm to predict the water content of the sugarcane plantation. The results are predicted water index values for the sugarcane plantation with 72% accuracy.

Systematic data-driven exploration of Austrian wastewater and sludge treatment - implications for phosphorus governance, costs and environment

Within the new policy framework shaped by the EU Green Deal and the Circular Economy Action Plans, the field of wastewater and sludge treatment in Europe is subject to high expectations and new challenges related to mitigation of greenhouse gas emissions, micropollutant removal and resource recovery. With respect to phosphorus recovery, several technologies and processes have been thoroughly investigated. Nevertheless, a systemic and detailed understanding of the existing infrastructure and of the related environmental and economic implications is missing. Such basis is essential to avoid unwanted consequences in designing new strategies, given the long lifespan of any infrastructural change. This study couples a newly collected and highly detailed database for all wastewater treatment plants in Austria bigger than 2000 population equivalent with a combination of analyses, namely Substance Flow Analysis with focus on nutrient and metal distribution in different environmental and anthropogenic compartments, Energy Flow Analysis, Life Cycle Assessment and cost estimation. The case study of Austria is of special interest, given its highly autonomous administration in federal states and its contrasting traits, ranging from flat metropolitan areas like Vienna to low-populated alpine areas. The significant impact of electricity demand of wastewater treatment on the overall Cumulative Energy Demand (CED) shows the importance of optimization measures. Further, the current system of wastewater and sludge disposal have a low efficiency in recovering nutrients and in directing pollutants as heavy metals into final sinks. Sludge composting with subsequent use in landscaping does not only show an unfavorable environmental balance, but it is the only relevant route leading to additional CED and Global Warming Potential emissions and to the highest transport volume. Altogether, the outcomes of this study provide a sound basis to further develop national strategies for resource recovery aimed to optimize trade-offs between different economic and environmental objectives.

Beyond traditional sleep scoring: Massive feature extraction and data-driven clustering of sleep time series

The widely used guidelines for sleep staging were developed for the visual inspection of electrophysiological recordings by the human eye. As such, these rules reflect a limited range of features in these data and are therefore restricted in accurately capturing the physiological changes associated with sleep. Here we present a novel analysis framework that extensively characterizes sleep dynamics using over 7700 time-series features from the hctsa software. We used clustering to categorize sleep epochs based on the similarity of their time-series features, without relying on established scoring conventions. The resulting sleep structure overlapped substantially with that defined by visual scoring. However, we also observed discrepancies between our approach and traditional scoring. This divergence principally stemmed from the extensive characterization by hctsa features, which captured distinctive time-series properties within the traditionally defined sleep stages that are overlooked with visual scoring. Lastly, we report time-series features that are highly discriminative of stages. Our framework lays the groundwork for a data-driven exploration of sleep sub-stages and has significant potential to identify new signatures of sleep disorders and conscious sleep states.

The factor structure of career indecision in China and South Korea: a data-driven exploration

The factor structure of career indecision in China and South Korea: a data-driven exploration

Towards a systematic understanding of blockchain governance in proposal voting: A dash case study

The transparent and immutable nature of blockchain provides incentives for organizations wishing to create and implement an open, decentralized governance structure. As members exercise their voting rights, a fault-tolerant record accumulates on the blockchain that can be analyzed to diagnose and intercept potential threats to the governing body. To date, there has not been a systematic study of on-chain governance with respect to voting. In this paper, we provide an analysis of blockchain governance through a case study of the first cryptocurrency to adopt on-chain voting, Dash. Our analysis introduces the key characteristics of blockchain governance, steps through a data-driven exploration of Dash's on-chain voting system, and highlights exploitable attack vectors and vulnerabilities for the subversion of Dash's on-chain voting system via a novel network analysis methodology. We then conclude with guidelines for other organizations looking to implement similar blockchain governance solutions while maintaining integrity in their operations.

A data-driven exploration of the race between human labor and machines in the 21 st century

To understand automation and the future of work, this study explores how human labor competes, or cooperates, with technology in performing a range of tasks.

Comparison of the Mechanical Properties and Microstructures of QB2.0 and C17200 Alloys.

As it is known, beryllium bronze, an important copper alloy, is widely used in the field of aerospace. Since the performance of domestic and imported beryllium bronze alloys have obvious differences, domestic beryllium bronze QBe2.0 and imported C17200 alloy were adopted, and the hardness and tensile properties of imported and domestic beryllium bronze alloys in the peak aging state were compared and analyzed. In addition, the microstructure morphologies of the C17200 alloy and QBe2.0 alloy were analyzed by SEM, EBSD, and TEM. This study adopted a data-driven exploration approach to elaborate the differences between C17200 and QBe2.0 alloy. After aging at 300 °C for 2 h (peak aging), the tensile strengths of the C17200 alloy and QBe2.0 alloy were 1357 MPa and 1309 MPa, the yield strengths were 1195 MPa and 1188 MPa, and the elongations were 5.5% and 4.0%, respectively. In the peak-aged state, the grain size, uniformity, small angle grain boundary, and twin density of the C17200 alloy were much better than those of the QBe2.0 alloy, which led to more significant grain refinement and twin strengthening effects. A large amount of γ’ phase, γ phase, and β phase were precipitated in both alloys, but the precipitation density of the γ’ strengthening phase in the C17200 alloy was much greater than that of the QBe2.0 alloy. The C17200 alloy exhibited better mechanical properties under the combined effects of the various strengthening mechanisms, which provided a guideline for the subsequent improvement of domestic alloys and laid a solid foundation for the development of new copper alloys.