Generation Of Digital Data Research Articles

On the Generation of Digital Data and Models from Point Clouds: Application to a Pedestrian Bridge Structure

This study investigates the capture of digital data and the development of models for structures with incomplete documentation and plans. LiDAR technology is utilized to obtain the point clouds of a pedestrian bridge structure. Two different point clouds with varying densities, (i) fine (11 collection locations) and (ii) coarse (4 collection locations), collected via terrestrial LiDAR, are analyzed to generate geometry and structural sections. This geometry is compared to the structural plans, which are then converted into numerical models (finite element—FE model) based on the point cloud data. Point cloud-based FE models (based on fine and coarse data) are compared with the structural plan-based FE model. It is observed that the static and dynamic responses are comparable within an acceptable range of a maximum difference of 5.5% for static deformation and an 8.23% frequency difference, with an average difference of less than 5%. Additionally, the dynamic properties of the fine and coarse point cloud FE models are compared with the operational modal analysis data obtained from the bridge. The fine and course point-cloud-based FE models, without model calibration, achieve an average accuracy of 8.76% and 9.94% for natural frequencies and a 0.89 modal assurance criterion value. The research found that the digital data generation yields promising results in this case for a bridge if documentation or plans are unavailable. With recent technologies and approaches such as digital twins, the connection between physical and virtual entities needs to be established by fusing digital models, sensorial information, and other data forms for better infrastructure management. Models such as those investigated and discussed in this paper can assist engineers with structural preservation in conjunction with monitoring data and utilization for digital twins.

A Novel Diabetes Prediction Model in Big Data Healthcare Systems Using DA-KNN Technique

In the past decades, there is a wide increase in the number of people affected by diabetes, a chronic illness. Early prediction of diabetes is still a challenging problem as it requires clear and sound datasets for a precise prediction. In this era of ubiquitous information technology, big data helps to collect a large amount of information regarding healthcare systems. Due to explosion in the generation of digital data, selecting appropriate data for analysis still remains a complex task. Moreover, missing values and insignificantly labeled data restrict the prediction accuracy. In this context, with the aim of improving the quality of the dataset, missing values are effectively handled by three major phases such as (1) pre-processing, (2) feature extraction, and (3) classification. Pre-processing involves outlier rejection and filling missing values. Feature extraction is done by a principal component analysis (PCA) and finally, the precise prediction of diabetes is accomplished by implementing an effective distance adaptive-KNN (DA-KNN) classifier. The experiments were conducted using Pima Indian Diabetes (PID) dataset and the performance of the proposed model was compared with the state-of-the-art models. The analysis after implementation shows that the proposed model outperforms the conventional models such as NB, SVM, KNN, and RF in terms of accuracy and ROC.

How Wearable Devices are Shaping the Insuretech Industry

The use of data has been revolutionizing every aspect of human life – both personal and professional. Every individual today is a source of digital data generation, and every business can benefit from access to such data if it is able to shape its strategies on the basis of insights derived from such data. Insurance has been a sector that has relied very heavily on mathematical modelling to devise policy plans that are appealing and beneficial to the target user groups as well as viable for the insurance provider. While such mathematical modelling relied on humongous data volumes, the growth of wearable devices has thrown up even more exciting possibilities ahead. Wearable technology data as a managed service is expected to have a size of $462 million globally in 2027. Wearable devices promise accessibility to live data of various body parameters of a huge population of users, which can help insurance companies build models to develop plans that have a greater utility and are more affordable. With a total of 533.6 million units shipped in 2021, representing a 20% growth over the previous year, we can be certain that as IoT and 5G technology improve, we will see far greater adoption of wearable devices. Along with aiding the development of more market-friendly insurance products, such devices can also help to a large extent in offering proactive healthcare.

Different Scales of Medical Data Classification Based on Machine Learning Techniques: A Comparative Study

In recent years, medical data have vastly increased due to the continuous generation of digital data. The different forms of medical data, such as reports, textual, numerical, monitoring, and laboratory data generate the so-called medical big data. This paper aims to find the best algorithm which predicts new medical data with high accuracy, since good prediction accuracy is essential in medical fields. To achieve the study’s goal, the best accuracy algorithm and least processing time algorithm are defined through an experiment and comparison of seven different algorithms, including Naïve bayes, linear model, regression, decision tree, random forest, gradient boosted tree, and J48. The conducted experiments have allowed the prediction of new medical big data that reach the algorithm with the best accuracy and processing time. Here, we find that the best accuracy classification algorithm is the random forest with accuracy values of 97.58%, 83.59%, and 90% for heart disease, M-health, and diabetes datasets, respectively. The Naïve bayes has the lowest processing time with values of 0.078, 7.683, and 22.374 s for heart disease, M-health, and diabetes datasets, respectively. In addition, the best result of the experiment is obtained by the combination of the CFS feature selection algorithm with the Random Forest classification algorithm. The results of applying RF with the combination of CFS on the heart disease dataset are as follows: Accuracy of 90%, precision of 83.3%, sensitivity of 100, and consuming time of 3 s. Moreover, the results of applying this combination on the M-health dataset are as follows: Accuracy of 83.59%, precision of 74.3%, sensitivity of 93.1, and consuming time of 13.481 s. Furthermore, the results on the diabetes dataset are as follows: Accuracy of 97.58%, precision of 86.39%, sensitivity of 97.14, and consuming time of 56.508 s.

Comparison of Deep Learning Models for Fully Automated Single Channel Idealization

Single ion channel analysis can reveal high resolution data about kinetics of single molecules, but has been described as “slow and laborious”. Typically, the first stage in such analysis is the creation of an idealized record with each timepoint classified as the number ion channels open. Whilst there have been a number of semi-automatic methods developed to assist with this, in situations with multiple channels in the patch or generally poor signal to noise ratio, these methods could be difficult to upscale these analysis for industrial use. With this in mind we are working on artificial intelligence methods that may one day supersede existing methods. One recent approach, called DeepChannel, used developments in deep learning (a branch of artificial intelligence) to fully idealize data fully automatically. Deep learning is heavily dependent on large datasets to train, which in DeepChannel's case were generated by passing simulated ion channel events through a real patch clamp amplifier. This method was thus limited to real-time recording resulting in the size of datasets to still be relatively small. Here, we describe new, completely digital data generation pipeline simulated data ion channel data and use these to train and compare new and old deep learning models including one based on the acclaimed image classification model “ResNet”. Additionally, we then use these models to idealise real world data, and compare with an expert annotation to judge relative model performance. Deep learning models show great potential for fully automated single molecule analysis and in the future we will release example models as a web app for easy use by other ion channel biologists.

Developing a supervised learning-based social media business sentiment index.

The fast-growing digital data generation leads to the emergence of the era of big data, which become particularly more valuable because approximately 70% of the collected data in the world comes from social media. Thus, the investigation of online social network services is of paramount importance. In this paper, we use the sentiment analysis, which detects attitudes and emotions toward issues of society posted in social media, to understand the actual economic situation. To this end, two steps are suggested. In the first step, after training the sentiment classifiers with several big data sources of social media datasets, we consider three types of feature sets: feature vector, sequence vector and a combination of dictionary-based feature and sequence vectors. Then, the performance of six classifiers is assessed: MaxEnt-L1, C4.5 decision tree, SVM-kernel, Ada-boost, Naïve Bayes and MaxEnt. In the second step, we collect datasets that are relevant to several economic words that the public use to explicitly express their opinions. Finally, we use a vector auto-regression analysis to confirm our hypothesis. The results show the statistically significant relationship between public sentiment and economic performance. That is, “depression” and “unemployment” lead to KOSPI. Also, it shows that the extracted keywords from the sentiment analysis, such as “price,” “year-end-tax” and “budget deficit,” cause the exchange rates.

The Impact of Digitization and Digital Data Mobilization on Biodiversity Research and Outreach

The first two decades of the 21st Century have seen a rapid rise in the creation, mobilization, research, and educational use of digital museum data, especially in the natural and biodiversity sciences. This has thrust natural history museums and especially the biodiversity specimen collections they hold into the forefront of biodiversity research in systematics, ecology, and conservation, underscoring their central role in the modern scientific enterprise. The advent of such digitization and data mobilization initiatives as the United States National Science Foundation’s Advancing the Digitization of Biodiversity Collections program (ADBC), Australia’s Atlas of Living Australia (ALA), Mexico’s National Commission for the Knowledge and Use of Biodiversity (CONABIO), Brazil’s Centro de Referência em Informação (CRIA), Europe’s SYNTHESYS, and China’s National Specimen Information Infrastructure (NSII) has led to a rapid rise in regional, national, and international digital data aggregators and has precipitated an exponential increase in the availability of digital data for scientific research. The international Global Biodiversity Information Facility (GBIF) now serves about 130 million museum specimen records, and Integrated Digitized Biocollections (iDigBio), the U.S. national biodiversity portal, has amassed over 109 million records representing over 300 million specimens that are international in scope. These digital resources raise the profiles of museums, expose collections to a wider audience of systematic and conservation researchers, provide the best biodiversity data in the modern era outside of nature itself, and ensure that specimen-based research remains at the forefront of the biodiversity sciences. Here we provide a brief overview of worldwide digital data generation and mobilization, the impact of these data on biodiversity research, new data underscoring the impact of worldwide digitization initiatives on citation in scientific publications, and evidence of the roles these activities play in raising the public and scientific profiles of natural history collections.

Big Data: Innovation and Competitive Advantage in an Information Media Analytics Company

The advent of technology has allowed for the capture of large volumes of data from a variety of sources. This has led to an ever-increasing number of firms collecting large amounts of data with the belief that this will give the firm an advantage over its competitors. However, the question is, does big data by itself really lead to firm advantage? And if not, how can firms gain an advantage from big data? This paper investigates the role that big data plays in innovation and firm advantage. Using the Resource-Based View and Dynamic Capabilities framework, this paper looks at how a firm can gain an advantage from big data. Through the analysis of a case-study concerning a global information/media analytics company, this paper provides an example of how to build a capability in Digital Data Generation that can lead to improved product or service innovation, and possibly help a firm gain an advantage over its competitors.

A Digital Grain Generation Method Suitable for Geometric Textures

The automated production of products using digital data has been realized in recent years through the development of manufacturing machines. Digital data are also used for automatic texturing. In a product texturing system using digital data, texture information needs to be incorporated in the generation of digital data. However, it is difficult to generate such data manually since it is not easy to fit the texture to the product or to prevent the texture pattern from being stretched or shortened. Therefore, what is needed is a method of automatically generating digital data that yields a seamless texture with little distortion and a method of fitting the texture to the product shape. In this paper, the application of the Angle Based Flattening (ABF) method by Sheffer and de Sturler to the generation of digital data of textures is proposed. Another method to extend the ABF technic to make it more suitable to geometric textures is also proposed.

GRAPHIC INTERPRETATION OF DIGITAL PRODUCTS AND APPLICATIONS IN STEGANOGRAPHY

Today, the generation of digital data is easier than it has ever been. These are the so-called digitally born products – documents, images, audio and video recordings, software, e-books, digital models, maps, etc. They are more and more widely employed nowadays in a variety of fields, such as business, science, art, education, digital communications, national security, intelligence service, military science, etc. Their global application has brought forth the need for their being reliably protected and has made it an issue of paramount importance in the modern world of information. This necessity has imposed the fast development of two contemporary sciences - cryptography and steganography. The present report treats one of the recent variations in the development of steganography - digital steganography, suggesting a universal idea for its application to all types of digital products.

The Data Problem in Data Mining

Computer science is essentially an applied or engineering science, creating tools. In Data Mining, those tools are supposed to help humans understand large amounts of data. In this position paper, I argue that for all the progress that has been made in Data Mining, in particular Pattern Mining, we are lacking insight into three key aspects: 1) How pattern mining algorithms perform quantitatively, 2) How to choose parameter settings, and 3) How to relate found patterns to the processes that generated the data. I illustrate the issue by surveying existing work in light of these concerns and pointing to the (relatively few) papers that have attempted to fill in the gaps. I argue further that progress regarding those questions is held back by a lack of data with varying, controlled properties, and that this lack is unlikely to be remedied by the ever increasing collection of real-life data. Instead, I am convinced that we will need to make a science of digital data generation, and use it to develop guidance to data practitioners.

Data centres in the UK: property and planning issues

PurposeThe continuing growth of digital technology has been accompanied by an explosion in digital data generation. These data are processed, stored, managed and exchanged in data centres but there is limited general awareness appreciation of the role such centres play as the driving hub of the economy. With this in mind, the purpose of this paper is to provide a simple review of the data centre property market and the changing geography of such centres and to review a number of the environmental and planning issues associated with the development of new data centres.Design/methodology/approachThe paper begins with a description of the characteristics and origins of data centres, outlines the data centre property market and the changing geography of new purpose built data centres within the UK and discusses a range of the environmental and planning issues associated with the development of such centres.FindingsThe paper reveals that four different types of data centre can be recognised, that relationship between supply and demand within the data centre property market is a contested issue and that the geography of new data centre development is changing. Data centre developers are increasingly looking for designs that increase energy efficiency and reduce carbon dioxide and the planning issues associated with such developments include classifying the use class, environmental and design considerations and employment generation.Research limitations/implicationsThe paper provides an accessible review of the development of data centres in the UK and as such it will be of interest to those working in, engaging with, and studying the data centre market.Originality/valueThe paper provides an accessible review of the development of data centres in the UK and as such it will be of interest to those working in, engaging with, and studying data centres.

The changing geography of data centres in the UK

ABSTRACTThe continuing growth of digital technology has been accompanied by an explosion in digital data generation. This data is processed, stored, managed and exchanged in data centres, which have become the driving hub of the economy and in some ways of society. This article outlines the origins and characteristics of data centres and offers an exploratory discussion of their locational dynamics and their changing geography within the UK. The findings reveal that while data centres were initially concentrated in London, a range of new urban, suburban and remote rural locations are becoming increasingly important. That said, while the geography of data centres may be changing the need to design ever more sustainable centres, which increase energy efficiency and reduce carbon emissions, seems likely to remain constant.

Digital ortho-images — a powerful tool for the extraction of spatial- and geo-information

Digital ortho-images provide great advantages in comparison to their analogue counterparts, especially with respect to flexibility, production of derived products and combination with other data sets. Today, production of digital ortho-images has become more commonplace due to the development of more powerful computers with sufficient resources, easier acquisition of input data, increased generation of digital data, development of many commercial ortho-image production systems, and new application areas, particularly in connection to GIS and digital mapping. After a review of past developments and the current status, the paper focuses on different extensions of ortho-images. These extensions permit the extraction of spatial information based on orthorectified stereo pairs, stereomates, monoplotting, and image analysis techniques. Emphasis is placed on the extraction of 3D information using orthorectified stereo pairs. An overview of practical usages of ortho-images in a GIS is also presented. Currently, digital ortho-images are primarily used either for visualization or as a backdrop. However, ortho-images offer many, to a large extent unexploited, possibilities, especially for the update and generation of databases, and automated mapping. The paper concludes with an outline of future trends in research and development.