Large Volume Datasets Research Articles

Environmental Sustainability Education in Twelve Series of Chinese University English Language Textbooks

Environmental issue is one of the public concerns pertaining to human welfare. English language education is supposed to take its part to contribute to the sustainable development. In this study, a corpus comprising 12 series of English language textbooks used in Chinese universities (48 volumes/books) was constructed to examine the extent to which the textbooks cover the content of environmental education, and how these topics are presented. The large volume dataset was explored by corpus linguistic tools and the in-depth discourse analysis and findings were reported and discussed within the theoretical framework of critical discourse analysis. The results suggested that some key environmental semantic domains, such as “Green issues” (e.g., nature, environment, ecology), “Substances and materials” (e.g., pesticide, chemical, carbon dioxide), “Science and technology” (e.g., technologies, nuclear, biotechnology), “Farming & Horticulture” (e.g., farming, crop, agriculture), “Living creatures” (e.g., endangered species, animals, wildlife) were evident. Some topics within these domains such as sustainable environmentalism, energy efficiency and sufficiency were presented and discussed critically. The textbooks are expected to help learners develop knowledge, competencies and futuristic attitudes needed to meet the sustainability challenges in various contexts.

A Classification Algorithm Inspired by the Chromatographic Separation Mechanism Dedicated to the Classification of Variable-length and Multi-class Vectors

Nowadays, one of the critical problems related to data mining is the processing of large data sets. This article presents an algorithm that may apply to the issues associated with classifying large-volume data sets. The motivation behind defining this type of algorithm was that the methods used to process this data type are subject to several significant limitations. The first considerable limitation of using classical classification methods is ensuring a constant data size. The second type of constraint is related to the data dimension. The last limitation in using classic classification algorithms is associated with the situation in which a given input vector may contain data belonging to many classes simultaneously, in which case we are talking about so-called multi-class vectors. The presented algorithm is inspired by the method of chromatographic separation of chemical substances. This method is widely and successfully used in analytical chemistry. As we know, in the case of chromatographic separation, we are dealing with a similar class of problems that occur when processing large data sets, firstly: the molecules of a chemical substance have a different number of molecules - i.e., they have different lengths, which corresponds to the situation that occurs when processing large data sets. In this work, a classification algorithm inspired by the mechanism of resolution chromatography is presented. The article presents the results of calculations for sample data sets. It discusses issues related to the properties of the defined algorithm, which concern the algorithm training process and the classification of single-class and multi-class data.

Managing large volume data sets in the process of identifying missing persons: Contributions from the International Commission on Missing Persons.

The process of locating and identifying missing persons presents a complex challenge that hinges on the collection and comparison of diverse data sets. This commentary offers an overview of some of the difficulties and considerations associated with data management in the context of large-scale missing person identifications. Such complexities include the uniqueness of each disaster event, the response time to the event, the variable quality and quantity of data, and the involvement of numerous stakeholders, all of which contribute to the intricacies of data management. In addition, the paramount considerations of privacy and ethical standards further compound these challenges, especially when dealing with sensitive information such as genetic data. This commentary describes the integrated Data Management System (iDMS) developed by the International Commission on Missing Persons (ICMP) as one example of a comprehensive, freely available solution for data collection, storage, protection, and analysis in missing person cases. The various advantages of the system are discussed, including the system's interoperability among the diverse array of stakeholders involved. While the iDMS streamlines data management processes and therefore represents a significant advancement in the field of missing person identification, it is concluded that the pending issue extends beyond thesoftware tools to encompass the lack of political will among stakeholders to collaborate there remains a pressing need for all stakeholders involved in the identification process to commit to a mechanism that facilitates compatibility and interoperability if different tools are used in disaster victim identification (DVI) scenarios.

FA -WSI -CNN Model for Predicting Breast Cancer using Deep Learning

Deep Learning is used for predicting a large volume of data sets in the medical field particularly for breast cancer prediction and diagnosis. The most effective and broadly applied model for detecting breast cancer is the Conventional Neural Network (CNN) among the various deep learning algorithms available. The existing CNN models are lacking in the analysis of a fully labeled Whole Set Image (WSI) data set. The proposed Fully Automate WSI with the CNN model will analyze the whole slide images and patch the input image for improving the accuracy. Then CNN model will get input from patched images and creates classified data for predicting breast cancer. The scikit-learn deep learning framework with Python is used to analyze the result and build a generalized tissue classifier, the WSI data set should include tissues generated under numerous different preparation circumstances. The proposed model experimental results shows promising WSI patch values, accuracy, precision, re-call, and F1 score of the breast cancer tissues which are used for diagnosis purposes. The FA -WSI -CNN model can reduce the training time by evaluating the inference time

Context-aware augmentation for liver lesion segmentation: shape uniformity, expansion limit and fusion strategy.

Data augmentation with context has been an effective way to increase the robustness and generalizability of deep learning models. However, to our knowledge, shape uniformity, expansion limit, and fusion strategy of context have yet to be comprehensively studied, particularly in lesion segmentation of medical images. To examine the impact of these factors, we take liver lesion segmentation based on the well-known deep learning architecture U-Net as an example and thoroughly vary the context shape, the expansion bandwidth as well as three representative fusion methods. In particular, the context shape includes rectangular, circular and polygonal, the expansion bandwidth is scaled by a maximum value of 2 compared to the lesion size, and the context fusion weighting strategy is composed of average, Gaussian and inverse Gaussian. Studies conducted on a newly constructed high-quality and large-volume dataset show that (I) uniform context improves lesion segmentation, (II) expanding the context with either 5 or 7 pixels yields the highest performance for liver lesion segmentation, depending on the lesion size, and (III) an unevenly distributed weighting strategy for context fusion is appreciated but in the opposite direction, depending on lesion size as well. Our findings and newly constructed dataset are expected to be useful for liver lesion segmentation, especially for small lesions.

Biomarkers for surgical sepsis. A review of foreign scientific and medical publications

remained unchanged for over a decade, and early recognition continues to be the most crucial factor in survival outcome. Early and accurate diagnosis of infection and organ dysfunction remains problematic, as evidenced by numerous interventional trials that have not resulted in improved outcomes. These failures are partly because of the belated intervention, when the patient developed multiple-organ failure and the therapeutic window of opportunity closed. The success of immunomodulatory and other therapeutic strategies, which is often achieved in preclinical models of sepsis, depends on their use in the early stages of sepsis development or even proactive action. Predicting the development of sepsis in surgical patients using laboratory analysis of plasma may be useful for doctors in the intensive care unit and resuscitation. Significant efforts are being made to develop biomarkers for the early stages of sepsis with high sensitivity and specificity. For early and accurate diagnosis, effective treatment of sepsis requires a deep understanding of the pathogenetic mechanisms. Dysregulation of the patients response to infection leading to sepsis and septic shock is studied using ohmic approaches: proteomics, transcriptomics, and metabolomics. Owing to the complexity and large volume of data sets, special data analysis tools, the so-called machine learning, become necessary.

Vo-Norvana : Versatile Framework for Efficient Segmentation of Large Point Cloud Data Sets

Dense three-dimensional (3D) point clouds collected from rapidly evolving data acquisition techniques such as light detection and ranging (lidar) and structure from motion (SfM) multiview stereo (MVS) photogrammetry contain detailed geometric information of a scene suitable for a wide variety of applications. Among the many processes within a typical point cloud processing workflow, segmentation is often a crucial step to group points with similar attributes to support more advanced modeling and analysis. Segmenting large point cloud data sets (i.e., hundreds of millions to billions of points) can be extremely time consuming and tedious with current tools, which primarily rely on significant manual effort. While many automated methods have been proposed, the practicality, scalability, and versatility of these approaches remain a bottleneck stifling processing of large data sets. To overcome these challenges, this paper introduces a novel, generalized segmentation framework called Vo-Norvana, which incorporates a new voxelization technique, a normal variation analysis considering the positioning uncertainty of the point cloud, and a custom region growing process for clustering. The proposed framework was tested with several large-volume data sets collected in diverse scene types using several data acquisition platforms including terrestrial lidar, mobile lidar, airborne lidar, and drone-based SfM-MVS photogrammetry. In evaluating the accuracy of models generated from Vo-Norvana against manual segmentation, the average error of the position, orientation, and dimensions are 2.7 mm, 0.083°, and 0.9 mm, respectively. Over 0.2 million points per second and 36 thousand voxels per second can be achieved when segmenting an airborne lidar data set containing over 639 million points to about 1 million segments.

Surface Crack Detection of Steel Structures in Railroad Industry Based on Multi-Model Training Comparison Technique

A method of steel structure surface crack identification based on artificial intelligence technology is proposed to solve the problem that steel cracks can not be detected and forewarned in time when they appear in the railway industrial environment. The appearance of steel cracks greatly weakens the stability of steel structures, and will seriously endanger the safety of the railway industry if it is not detected and repaired in time. However, the common steel crack detection methods cannot achieve real-time monitoring of steel structures. In order to monitor the surface of steel structure in real-time and explore the recognition effect and model the advantages of common classification neural network models for surface cracks of railway industrial steel, this study evaluates the network model with multiple indicators and parameters under two experimental conditions. In this study, the steel surface cracks in the railway industrial environment are taken as samples, and the steel cracks are identified through the neural network model. For large-volume datasets, the recognition accuracy of the three network models has reached 97%, of which the YOLOv5 model has the best comprehensive recognition ability, and the C-Alex model has the best performance and convergence speed in small-volume datasets. This study explores the application prospects of models under different scenarios, proving that the three models can effectively detect steel surface cracks in real-time, and at the same time, it will pave the way for the development and application of artificial intelligence multi-model fusion technology in the field of the railway industry.

Application of deep learning to support peak picking during non-target high resolution mass spectrometry workflows in environmental research

A CNN was developed to classify extracted features from nontarget mass spectrometry workflows. The CNN accuracy ranged from 85% to 100%. These tools will be important in data-driven research enabling rapid processing of large volume and complex datasets.

A Novel Approach to Improve the Performance of the Database Storing Big Data with Time Information

Big data is defined as data sets that are too large and/or complex to be processed by classical data processing methods. Big data analysis is essential because it enables more competent business movements, more efficient operations, and higher profits by using the data of institutions and organizations. However, large datasets are difficult to analyze because they are produced quickly, require large storage areas in computer systems, and the diversity of their data. In this study, a new approach using the denormalization method is proposed to accelerate the response time of the database in database systems where large volumes of data containing historical information are stored. Denormalization is defined as the process of adding rows or columns that are not needed to increase the reading performance of the database to the database system that has been normalized. In the proposed approach in this study, a large-volume dataset consisting of real spatial data belonging to Kayseri Metropolitan Municipality, containing temporal information and having approximately 96,000,000 row records, was used. In the proposed approach, the response time of the query is accelerated by recording the time information as numbers to increase the query performance of large volumes of data recorded in date format due to the temporal query process. The performance of the proposed method is compared with the performance of the normalization method using actual data on Microsoft SQL Server and Oracle database systems. The method proposed in the experimental evaluations shows that it works approximately eight times faster. In addition, the experimental results showed that the proposed method improves query performance more than the normalization-based method as the data size increases.

Facing the Challenges in simulation-based Earth System Sciences and the Role of FAIR Digital Objects

Facing the Challenges in simulation-based Earth System Sciences and the Role of FAIR Digital Objects

A Comprehensive Analysis Of Normalization Approaches For Privacy Protection In Data Mining

Data Mining is a fundamental method of extracting large volumes of data sets by unfamiliar patterns. These extracted data can be shared between enterprises to improve their corporate benefits. For the data mining process, sharing such confidential data is very important. It is very important to safeguard such information against unwanted exposure that leads to privacy leakage. In recent days, privacy in various data mining applications has become very important. In order to overcome such problem privacy, data mining techniques are preserved. It provides accurate data mining results without scarifying the original data values and ensures both accuracy and privacy. Data protection is achieved through the analysis of the data using normalization techniques in this proposed work. The approach proposed in comparison the effects of min-max, decimal scaling, and Z-Score normalization techniques. Experimental effects confirmed that the Min-max Standardization technology archived the most precision with minimum data loss.

An XAI method for convolutional neural networks in self-driving cars.

eXplainable Artificial Intelligence (XAI) is a new trend of machine learning. Machine learning models are used to predict or decide something, and they derive output based on a large volume of data set. Here, the problem is that it is hard to know why such prediction was derived, especially when using deep learning models. It makes the models unreliable in the case of reliability-critical applications. So, it is required to explain how they derived such output. It is a reliability-critical application for self-driving cars because the mistakes made by the computers inside them can lead to critical accidents. So, it is necessary to adopt XAI models in this field. In this paper, we propose an XAI method based on computing and explaining the difference of the output values of the neurons in the last hidden layer of convolutional neural networks. First, we input the original image and some modified images of it. Then we derive output values for each image and compare these values. Then, we introduce the Sensitivity Analysis technique to explain which parts of the original image are needed to distinguish the category. In detail, we divide the image into several parts and fill these parts with shades. First, we compute the influence value on the vector indicating the last hidden layer of the model for each of these parts. Then we draw shades whose darkness is in proportion to the influence values. The experimental results show that our approach for XAI in self-driving cars finds the parts needed to distinguish the category of these images accurately.

Set Representation for Rule Generation Algorithms

The task of mining the association rule has become one of the most widely used discovery pattern methods in Knowledge Discovery in Databases (KDD). One such task is to represent the itemset in the memory. The representation of the itemset largely depend on the type of data structure that is used for storing them. Computing the process of mining the association rule im- pacts the memory and time requirement of the itemset. With the increase in the dimensionality of data and datasets, mining such large volume of datasets will be difficult since all these itemsets cannot be placed in the main memory. As representation of an itemset greatly affects the efficiency of the rule mining association, a compact and compress representation of an itemset is needed. In this paper, a set representation is introduced which is more memory and cost efficient. Bitmap representation takes one byte for an element but the set representation uses one bit. The set representation is being incorporated in Apriori Algorithm. Set representation is also being tested for different rule generation algorithms. The complexities of these different rule generation algorithms using set representation are being compared in terms of memory and time execution.

Efficacy of Deep Learning Models in the Prognosis Task

Deep learning model is an improved version of machine learning technology which brings out the efficiency much better on multivariate data. Deep Leaning models are performed actively and powerfully than current machine learning models. It triggers computational model to learn automatically from the given data at multiple levels. The reputation of deep learning is progressed on today’s large volume of data set and the advancement of hardware that provides potential computing systems. This article analyzes various deep learning models based on the architecture, activation functions, merits, and demerits in a novel way.

Utilising Big Data Analytics in the E-Government Health Sector

The world has witnessed massive technological improvements in the Information Technology (IT) sector, giving rise to unprecedented IT penetration into every facet of life and daily living. It has affected the way we live and interact, the way businesses are conducted, and most especially how the government makes decisions that affect its citizenry. Presently, government officials deploy IT techniques in their decision-making process to reach the ever-growing needs of their populace. The traditional approach to large data processing is proving difficult. Hence, the dawn of Big Data Analytics, a phenomenon that describes the processing of large volumes of datasets, being generated at a very high velocity, coming in various forms, and are largely unstructured. This paper examines literatures on Big Data Analytics and its application in e-government. A prototype framework which is divided into two will be proposed for its application. The first section will have the Hadoop infrastructure deployed for distributed storage in clusters, while the second section used a machine learning software – Waikato Environment for Knowledge Analysis (WEKA) for data mining. Finally, the different data mining algorithms provided by WEKA was explored and used in analyzing medical records obtained from the UCI online repository to demonstrate the data analysis of the proposed prototype. From the results of the analysis, J48 algorithm was used to build prediction trees, to ascertain patterns that determine the likelihood of citizens having breast cancer, and to generate predictions rules that will help in curbing or detecting breast cancer early among the populace.

ProvNet: Networked bi-directional blockchain for data sharing with verifiable provenance

Data sharing is increasingly popular especially for scientific research and business fields where large volume of datasets need to be used, but it involves data security and privacy concerns. This paper mitigates such concerns by tracking and logging the history of shared data (i.e., provenance records) while preserving data privacy. This is a challenging problem in the data sharing scenario in this paper because the environment is decentralized and internal logs are not accessible publicly due to privacy concerns. We present ProvNet, a decentralized data sharing platform that can detect malicious users and provide secure provenance records using the newly proposed networked blockchain without disclosing raw data contents. Valid sharing records are collected and stored in the blocknet and misbehavior is detected with the stored provenance records according to our accountable protocols. We give a proof-of-concept implementation, and evaluation results show that the overhead is acceptable.

Data Mining Approach For Exploring Soil Fertility

Predication of crop is a widespread problem . Farmer always had a curiosity to know his yield production, climatic condition, how the yield will be better. Simplifying the agriculture production is the essentiality of agriculture improvement. There is need to develop good techniques for crop prediction and other related aspects. Data mining has been emerged a great field in the agriculture. Data mining enables farmers to identify potentially interesting and unknown patterns in large volume of datasets. This paper discuss about the techniques of data mining that helps the farmers to overcome the problems and provide a suitable solution to them. Keywords:- Data mining , Clustering , Bi -Clustering , K-Nearest Method , Support Vector Machines(SVMs)

Image-based Visualization of Large Volumetric Data Using Moments.

We present a novel image-based representation to interactively visualize large and arbitrarily structured volumetric data. This image-based representation is created from a fixed view and models the scalar densities along each viewing ray. Then, any transfer function can be applied and changed interactively to visualize the data. In detail, we transform the density in each pixel to the Fourier basis and store Fourier coefficients of a bounded signal, i.e. bounded trigonometric moments. To keep this image-based representation compact, we adaptively determine the number of moments in each pixel and present a novel coding and quantization strategy. Additionally, we perform spatial and temporal interpolation of our image representation and discuss the visualization of introduced uncertainties. Moreover, we use our representation to add single scattering illumination. Lastly, we achieve accurate results even with changes in the view configuration. We evaluate our approach on two large volume datasets and a time-dependent SPH dataset.

Bioinformatics in Lipidomics: Automating Large-Scale LC-MS-Based Untargeted Lipidomics Profiling with SimLipid Software.

Liquid chromatography-mass spectrometry (LC-MS) provides one of the most popular platforms for untargeted plant lipidomics analysis (Shulaev and Chapman, Biochim Biophys Acta 1862(8):786-791, 2017; Rupasinghe and Roessner, Methods Mol Biol 1778:125-135, 2018; Welti et al., Front Biosci 12:2494-506, 2007; Shiva et al., Plant Methods 14:14, 2018). We have developed SimLipid software in order to streamline the analysis of large-volume datasets generated by LC-MS-based untargeted lipidomics methods. SimLipid contains a customizable library of lipid species; graphical user interfaces (GUIs) for visualization of raw data; the identified lipid molecules and their associated mass spectra annotated with fragment ions and parent ions; and detailed information of each identified lipid species all in a single workbench enabling users to rapidly review the results by examining the data for confident identifications of lipid molecular species. In this chapter, we present the functionality of the software and workflow for automating large-scale LC-MS-based untargeted lipidomics profiling.