Data Analysis Framework Research Articles

Public Health Impacts on Economic Growth in Developing Countries: An Analytical Review of Research

This paper identifies an analytical review of the study on how public health impacts economic growth in developing countries of South Asia such as Afghanistan, Bangladesh, Bhutan, India, Nepal, Pakistan and Sri Lanka by using the data from 2009 to 2023 by WDI. This research used the panel data framework for data analysis. Overall panel regression finding shows that the effect of public health such as capital stock and life expectancy on sustainable economic growth is positive. Furthermore, fertility rate and mortality rate negative impact on GDP or sustainable economic growth in South Asia region of developing countries. These findings suggest that sustainable economic growth can be improved by increasing public health facilities as a result, sustainable economic growth improves. The study also suggested recommendations for policy options and innovative solutions that lead to improved economic growth.

Protect your data and I’ll rank its utility: A framework for utility analysis of anonymized mobility data for smart city applications

When designing smart cities’ building blocks, mobility data plays a fundamental role in applications and services. However, mobility data usually comes with unrestricted location of its corresponding entities (e.g., citizens and vehicles) and poses privacy concerns, among them recovering the identity of those entities with linking attacks. Location Privacy Protection Mechanisms (LPPMs) based on anonymization, such as mix-zones, have been proposed to address the privacy of users’ identity. Once the data is protected, a comprehensive discussion about the trade-off between privacy and utility happens. However, issues still arise about the application of anonymized data to smart city development: what are the smart cities applications and services that can best leverage mobility data anonymized by mix-zones? To answer this question, we propose the Utility Analysis Framework of Anonymized Trajectories for Smart Cities-Application Domains (UAFAT). This characterization framework measures the utility through twelve metrics related to privacy, mobility, and social, including mix-zones performance metrics from anonymized trajectories produced by mix-zones. This framework aims to identify applications and services where the anonymized data will provide more or less utility in various aspects. The results evaluated with cabs and privacy cars datasets showed that further characterizing it by distortion level, UAFAT ranked the smart cities application domains that best leverage mobility data anonymized by mix-zones. Also, it identified which one of the four case studies of smart city applications had more utility. Additionally, different datasets present different behaviors in terms of utility. These insights can contribute significantly to the utility of both open and private data markets for smart cities.

HyperGCN: an effective deep representation learning framework for the integrative analysis of spatial transcriptomics data

BackgroundAdvances of spatial transcriptomics technologies enabled simultaneously profiling gene expression and spatial locations of cells from the same tissue. Computational tools and approaches for integration of transcriptomics data and spatial context information are urgently needed to comprehensively explore the underlying structure patterns. In this manuscript, we propose HyperGCN for the integrative analysis of gene expression and spatial information profiled from the same tissue. HyperGCN enables data visualization and clustering, and facilitates downstream analysis, including domain segmentation, the characterization of marker genes for the specific domain structure and GO enrichment analysis.ResultsExtensive experiments are implemented on four real datasets from different tissues (including human dorsolateral prefrontal cortex, human positive breast tumors, mouse brain, mouse olfactory bulb tissue and Zabrafish melanoma) and technologies (including 10X visium, osmFISH, seqFISH+, 10X Xenium and Stereo-seq) with different spatial resolutions. The results show that HyperGCN achieves superior clustering performance and produces good domain segmentation effects while identifies biologically meaningful spatial expression patterns. This study provides a flexible framework to analyze spatial transcriptomics data with high geometric complexity.ConclusionsHyperGCN is an unsupervised method based on hypergraph induced graph convolutional network, where it assumes that there existed disjoint tissues with high geometric complexity, and models the semantic relationship of cells through hypergraph, which better tackles the high-order interactions of cells and levels of noise in spatial transcriptomics data.

The Operational Data Analytics (ODA) for Service Speed Design

We develop the operational data analytics (ODA) framework for the classical service design problem of [Formula: see text] systems. The customer arrival rate is unknown. Instead, some historical data of interarrival times are collected. The data-integration model, specifying the mapping from the arrival data to the service rate, is formulated based on the time-scaling property of the stochastic service process. Validating the data-integration model against the long-run average service reward leads to a uniformly optimal service rate for any given sample size. We further derive the ODA-predicted reward function based on the data-integration model, which gives a consistent estimate of the underlying reward function. Our numerical experiments show that the ODA framework can lead to an efficient design of service rate and service capacity, which is insensitive to model specification. The ODA solution exhibits superior performance compared with the conventional estimation-and-then-optimization solutions in the small sample regime. This paper was accepted by David Simchi-Levi, operations management. Funding: Z. Jiang’s research is supported by the National Natural Science Foundation of China [Grant 71931007]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2023.00655 .

A Generic Integrated Framework of Unsupervised Learning and Natural Language Processing Techniques for Digital Healthcare: A Comprehensive Review and Future Research Directions

Abstract:: The increasing availability of digital healthcare data has opened up fresh prospects for improving healthcare through data analysis. Machine learning (ML) procedures exhibit great promise in analyzing large volumes of healthcare data to extract insights that could be utilized to improve patient outcomes and healthcare delivery. In this work, we suggest an integrated framework for digital healthcare data analysis by integrating unsupervised learning techniques and natural language processing (NLP) techniques into the analysis pipeline. The module on unsupervised learning will involve techniques, such as clustering and anomaly detection. By clustering similar patients together based on their medical history and other relevant factors, healthcare providers can identify subgroups of patients who may require different treatment approaches. Anomaly detection can also help to detect patients who stray from the norm, which could be indicative of underlying health issues or other issues that need additional investigation. The second module on NLP will enable healthcare providers to analyze unstructured text data such as clinical notes, patient surveys, and social media posts. NLP techniques can help to identify key themes and patterns in these datasets, requiring awareness that could not be readily apparent through other means. Overall, incorporating unsupervised learning techniques and NLP into the analysis pipeline for digital healthcare data possesses the promise to enhance patient results and lead to more personalized treatments, and represents a potential domain for upcoming research in this field. In this research, we also review the current state of research in digital healthcare information examination with ML, including applications like forecasting clinic readmissions, finding cancerous tumors, and developing personalized drug dosing recommendations. We also examine the potential benefits and challenges of utilizing ML in healthcare data analysis, including issues related to data quality, privacy, and interpretability. Lastly, we discuss the forthcoming research paths, involving the necessity for enhanced methods for incorporating information from several resources, developing more interpretable ML patterns, and addressing ethical and regulatory challenges. The usage of ML in digital healthcare data analysis promises to transform healthcare by empowering more precise diagnoses, personalized treatments, and improved health outcomes, and this work offers a complete overview of the current trends.

Impact of beam far side-lobe knowledge in the presence of foregrounds for LiteBIRD

We present a study of the impact of a beam far side-lobe lack of knowledge on the measurement of the Cosmic Microwave Background B-mode signal at large scale. Beam far side-lobes induce a mismatch in the transfer function of Galactic foregrounds between the dipole and higher multipoles which degrads the performances of component separation methods. This leads to foreground residuals in the CMB map. It is expected to be one of the main source of systematic effects in future CMB polarization observations. Thus, it becomes crucial for all-sky survey missions to take into account the interplays between beam systematic effects and all the data analysis steps. LiteBIRD is the ISAS/JAXA second strategic large-class satellite mission and is dedicated to target the measurement of CMB primordial B modes by reaching a sensitivity on the tensor-to-scalar ratio r of σ(r) ≤ 10-3 assuming r = 0. The primary goal of this paper is to provide the methodology and develop the framework to carry out the end-to-end study of beam far side-lobe effects for a space-borne CMB experiment. We introduce uncertainties in the beam model, and propagate the beam effects through all the steps of the analysis pipeline, most importantly including component separation, up to the cosmological results in the form of a bias δr. As a demonstration of our framework, we derive requirements on the calibration and modeling for the LiteBIRD's beams under given assumptions on design, simulation, component separation method and allocated error budget. In particular, we assume a parametric method of component separation with no mitigation of the far side-lobes effect at any stage of the analysis pipeline. We show that δr is mostly due to the integrated fractional power difference between the estimated beams and the true beams in the far side-lobes region, with little dependence on the actual shape of the beams, for low enough δr. Under our set of assumptions, in particular considering the specific foreground cleaning method we used, we find that the integrated fractional power in the far side-lobes should be known at the level of ∼ 10-4, to achieve the required limit on the bias δr < 1.9 × 10-5. The framework and tools developed for this study can be easily adapted to provide requirements under different design, data analysis frameworks and for other future space-borne experiments, such as PICO or CMB-Bharat. We further discuss the limitations of this framework and potential extensions to circumvent them.

Machine Learning-based Distributed Big data Analytics Framework for IoT Applications

The emerging trend of the Internet of Things (IoT) leads to more real-time applications and it raises the accumulation of more structured and unstructured data. The processing of unstructured and structured data for different distributed applications becomes arduous. Moreover, the accuracy, load balancing, and latency of the services are also challenging. Some of the state-of-art works failed to achieve those parameters. In context with these, we proposed a machine learning-based novel approach that utilizes an SVM classifier. The SVM classifier can be used for the classification and data analytic purposes of features extracted from the data processing step. Data processing employs two steps such as data extraction and data scaling. The data extraction is performed by the adoption of the Principal Component Discriminant power-based Linear Discriminant analysis (PCDP-LDA) technique. The big data framework of the proposed model is evaluated using a tool called Weka. Meanwhile, the data scaling is performed by the Naïve Bayes approach and divides the extracted features into blocks. Experimental analysis is performed and compared with existing approaches. Our proposed approach provides more effective classification and data analytic accuracy than the other approaches. Our approach also provides better latency, and load balancing of data in the distributed big data analysis.

CASi: A framework for cross-timepoint analysis of single-cell RNA sequencing data

Single-cell RNA sequencing (scRNA-seq) technology has been widely used to study the differences in gene expression at the single cell level, providing insights into the research of cell development, differentiation, and functional heterogeneity. Various pipelines and workflows of scRNA-seq analysis have been developed but few considered multi-timepoint data specifically. In this study, we develop CASi, a comprehensive framework for analyzing multiple timepoints’ scRNA-seq data, which provides users with: (1) cross-timepoint cell annotation, (2) detection of potentially novel cell types emerged over time, (3) visualization of cell population evolution, and (4) identification of temporal differentially expressed genes (tDEGs). Through comprehensive simulation studies and applications to a real multi-timepoint single cell dataset, we demonstrate the robust and favorable performance of the proposal versus existing methods serving similar purposes.

Corporate Governance Dynamics and Financial Performance: Analysis of Listed Commercial Banks in the Ghanaian Context"

This study aims to determine the effect of corporate governance on the financial performance of 14 listed commercial banks in Ghana. The research employs a panel data set from 2008 to 2020 to examine how board size, ownership structure, board independence, capital adequacy ratio, audit committee effectiveness, board gender diversity, and overall corporate governance score affect ROA and ROE for banks. These relationships are analyzed using Ordinary Least Squares (OLS) regression within the framework of panel data analysis. Results indicate that ROA is positively related to board size, board independence, audit committee, and Board gender diversity. Also, ROE is positively influenced by companies’ ownership structure, capital adequacy ratio, audit committee effectiveness, and overall corporate employment governance rating. However, there is a significant negative correlation between ROE and women’s representation or board gender diversity. Also, another key finding is that both Return on Assets (ROA) and Return on Equity (ROE) are significantly affected by debt and debt-to-equity ratios. This indicates that banks borrowing more have reduced profitability, underlining the need for a well-managed capital structure and no over-reliance on loans. This study contributes to the understanding of the relationship between corporate governance and performance in the banking sector of Ghana, emphasizing that high-quality governance systems can improve and enhance financial performance. The insight gained from this analysis shows that Banks should aim towards good corporate governance practices such as; maintaining proper board size, promoting greater board independence, ensuring efficient audit procedures, and encouraging more gender diversity to enhance profitability. The outcomes also emphasized a well-built base of capital and concentrated ownership, enabling better financial results.The outcomes of this research are useful to Ghanaian policymakers, regulators, and bank executives who view the need for strong corporate governance environments to buttress the stability and growth of the banking sector.

SHARE: A Framework for Secondary Qualitative Data Analysis

SHARE: A Framework for Secondary Qualitative Data Analysis

Density estimation via measure transport: Outlook for applications in the biological sciences

AbstractOne among several advantages of measure transport methods is that they allow or a unified framework for processing and analysis of data distributed according to a wide class of probability measures. Within this context, we present results from computational studies aimed at assessing the potential of measure transport techniques, specifically, the use of triangular transport maps, as part of a workflow intended to support research in the biological sciences. Scenarios characterized by the availability of limited amount of sample data, which are common in domains such as radiation biology, are of particular interest. We find that when estimating a distribution density function given limited amount of sample data, adaptive transport maps are advantageous. In particular, statistics gathered from computing series of adaptive transport maps, trained on a series of randomly chosen subsets of the set of available data samples, leads to uncovering information hidden in the data. As a result, in the radiation biology application considered here, this approach provides a tool for generating hypotheses about gene relationships and their dynamics under radiation exposure.

Machine Learning-based Distributed Big data Analytics Framework for IoT Applications

The emerging trend of the Internet of Things (IoT) leads to more real-time applications and it raises the accumulation of more structured and unstructured data. The processing of unstructured and structured data for different distributed applications becomes arduous. Moreover, the accuracy, load balancing, and latency of the services are also challenging. Some of the state-of-art works failed to achieve those parameters. In context with these, we proposed a machine learning-based novel approach that utilizes an SVM classifier. The SVM classifier can be used for the classification and data analytic purposes of features extracted from the data processing step. Data processing employs two steps such as data extraction and data scaling. The data extraction is performed by the adoption of the Principal Component Discriminant power-based Linear Discriminant analysis (PCDP-LDA) technique. The big data framework of the proposed model is evaluated using a tool called Weka. Meanwhile, the data scaling is performed by the Naïve Bayes approach and divides the extracted features into blocks. Experimental analysis is performed and compared with existing approaches. Our proposed approach provides more effective classification and data analytic accuracy than the other approaches. Our approach also provides better latency, and load balancing of data in the distributed big data analysis.

Multimodal Moore-Penrose Inverse-Based Recomputation Framework for Big Data Analysis.

Most multilayer Moore-Penrose inverse (MPI)-based neural networks, such as deep random vector functional link (RVFL), are structured with two separate stages: unsupervised feature encoding and supervised pattern classification. Once the unsupervised learning is finished, the latent encoding is fixed without supervised fine-tuning. However, in complex tasks such as handling the ImageNet dataset, there are often many more clues that can be directly encoded, while unsupervised learning, by definition, cannot know exactly what is useful for a certain task. There is a need to retrain the latent space representations in the supervised pattern classification stage to learn some clues that unsupervised learning has not yet been learned. In particular, the residual error in the output layer is pulled back to each hidden layer, and the parameters of the hidden layers are recalculated with MPI for more robust representations. In this article, a recomputation-based multilayer network using Moore-Penrose inverse (RML-MP) is developed. A sparse RML-MP (SRML-MP) model to boost the performance of RML-MP is then proposed. The experimental results with varying training samples (from 3k to 1.8 million) show that the proposed models provide higher Top-1 testing accuracy than most representation learning algorithms. For reproducibility, the source codes are available at https://github.com/W1AE/Retraining.

APPLICATION OF ARTIFICIAL INTELLIGENCE FOR SOLUTION OF ENGINEERING PROBLEMS. ADVANTAGES AND CHALLENGES

The intensive development of information technology leads to an increase in data volumes, which requires the application of effective methods for data processing and analysis for managing organizations and strategic planning. Data mining methods (DM) are widely used in various fields, including education, where they can help improve the quality of learning and the efficiency of teachers. This work demonstrates the use of the Orange platform, a framework for data visualization and analytics, which allows integrating visual programming with Python to solve complex analytical tasks. The application of Data Mining methods and the use of Orange allow for a deep analysis of educational data, contributing to the development of strategies to improve the efficiency of the learning process.

Large-scale automated emission measurement of individual vehicles with point sampling

Abstract. Currently, emissions from internal combustion vehicles are not properly monitored throughout their life cycle. In particular, a small share of vehicles (&lt; 20 %) with malfunctioning after-treatment systems and old vehicles with outdated engine technology are responsible for the majority (60 %–90 %) of traffic-related emissions. Remote emission sensing (RES) is a method used for screening emissions from a large number of in-use vehicles. Commercial open-path RES systems are capable of providing emission factors for many gaseous compounds, but they are less accurate and reliable for particulate matter (PM). Point sampling (PS) is an extractive RES method where a portion of the exhaust is sampled and then analyzed. So far, PS studies have been predominantly conducted on a rather small scale and have mainly analyzed heavy-duty vehicles (HDVs), which have high exhaust flow rates. In this work, we present a comprehensive PS system that can be used for large-scale screening of PM and gas emissions, largely independent of the vehicle type. The data analysis framework developed here is capable of processing data from thousands of vehicles. The core of the data analysis is our peak detection algorithm (TUG-PDA), which determines and separates emissions down to a spacing of just a few seconds between vehicles. We present a detailed evaluation of the main influencing factors on PS measurements by using about 100 000 vehicle records collected from several measurement locations, mainly in urban areas. We show the capability of the emission screening by providing real-world black carbon (BC), particle number (PN) and nitrogen oxide (NOx) emission trends for various vehicle categories such as diesel and petrol passenger cars or HDVs. Comparisons with open-path RES and PS studies show overall good agreement and demonstrate the applicability even for the latest Euro emission standards, where current open-path RES systems reach their limits.

All the better for it: Exploring one teacher-researcher's evolving efforts to promote meaningful physical education

Drawing on social constructivist learning theory and a conceptual framework for meaningful physical education (MPE), the purpose of this self-study was to present and explore how one teacher-researcher (TR) evolved their approach to implementing pedagogy promoting MPE in an alternative secondary/high school education setting. Utilizing self-study, data were collected before, during, and after twenty 75-minute lessons over 15 weeks. One critical friend interview, 20 post-teaching reflections, 18 observations, and 22 journal entries were conducted. A deductive and inductive approach utilizing the Miles, Huberman and Saldana Framework for Qualitative Data Analysis (2014) was implemented, with thematic analysis then applied. The following themes were constructed: bringing the fun; steering between sparkling and sensitive social interactions; co-creating and reconsidering challenge; monitoring movements toward motor competence; and pursuing personal relevance in physical activity. Findings demonstrate the TR's explicit and intentional efforts to better implement the concept of MPE within their planning and pedagogy to develop a better understanding of what this looked like in practice and the role each feature played in teaching and learning. Utilizing and promoting democratic and reflective practices led the TR to consider and apply these features more frequently and readily to better teach knowledge, attitudes, and skills to their students through the conceptual lens of MPE. The experiences presented and discussed demonstrate the benefit of doing so not just for the TR's evolving approach to teaching and learning in physical education, but indeed for the learning of their students, researchers, and other practitioners too.

Development of sensor system and data analytic framework for non-invasive blood glucose prediction

AbstractPeriodic quantification of blood glucose levels is performed using painful, invasive methods. The proposed work presents the development of a noninvasive glucose-monitoring device with two sensors, i.e., finger and wrist bands. The sensor system was designed with a near-infrared (NIR) wavelength of 940 nm emitter and a 900–1700 nm detector. This study included 101 diabetic and non-diabetic volunteers. The obtained dataset was subjected to pre-processing, exploratory data analysis (EDA), data visualization, and integration methods. Ambiguities such as the effects of skin color, ambient light, and finger pressure on the sensor were overcome in the proposed ‘niGLUC-2.0v’. niGLUC-2.0v was validated with performance metrics where accuracy of 99.02%, mean absolute error (MAE) of 0.15, mean square error (MSE) of 0.22 for finger, and accuracy of 99.96%, MAE of 0.06, MSE of 0.006 for wrist prototype with ridge regression (RR) were achieved. Bland–Altman analysis was performed, where 98% of the data points were within ± 1.96 standard deviation (SD), 100% were under zone A of the Clarke Error Grid (CEG), and statistical analysis showed p &lt; 0.05 on evaluated accuracy. Thus, niGLUC-2.0v is suitable in the medical and personal care fields for continuous real-time blood glucose monitoring.

The Data Analytics Framework for XDMoD

The Data Analytics Framework for XDMoD

Distributed Systems for Data-Intensive Computing in Cloud Environments: A Review of Big Data Analytics and Data Management

Because of the increasing increase of data, which is frequently referred to as "big data," many different businesses have been severely impacted in recent years, necessitating the implementation of sophisticated data management and analytics solutions. By virtue of the fact that it provides scalable resources for applications that are data-intensive, cloud computing has emerged as an indispensable platform for the management of these enormous databases. The evolving landscape of distributed systems in cloud settings is the primary emphasis of this study, which is situated within the framework of big data analytics and data management. With the purpose of providing a comprehensive overview of distributed systems that are used in cloud settings for data-intensive computing, the review article seeks to offer. Furthermore, it evaluates the many ideas, techniques, and technical improvements that have been established in order to properly manage, store, and analyse large amounts of data. A comprehensive literature evaluation of recently published scientific references was successfully completed by our team. The analysis takes into account the theoretical foundations, as well as the research that has already been conducted on distributed computing systems, cloud-based data management, and enormous data analytics. The study places an emphasis on the significant role that distributed computing plays in ensuring the success of big data analytics. The interplay between distributed systems and cloud computing paradigms has resulted in the development of solutions that are robust, scalable, and economical for activities that need a significant amount of data. It is still a huge problem to ensure that data security, privacy, and interoperability are maintained across the many cloud services.

Big Data Analysis and Deep Learning Optimization in Artificial Intelligence Production of Information Enterprises

Intelligent manufacturing technology is required to upgrade existing enterprises' management and production operations. To construct a ground breaking fusion structure, this project unites the theoretical underpinnings, technical breakthroughs, and applications of data analytics, optimisation, and intelligent production engineering. It is driven by China's desire of cutting-edge commodities and efficient growth methods. This research establishes the broad framework for merging optimisation and data analytics. There is a list of data analytics and system optimisation technologies that can address important challenges with intelligent manufacturing. By integrating data analytics and optimisation, businesses may better forecasting and management of new terrain, as well as reveal hidden information to increase decision-making efficacy.