Big Data Analysis Methods Research Articles

Satellite wave 2D spectrum partition based on the PI-vit-GAN(physically-informed ViT-GAN) method

The abundant spectral data provided by satellite technology are crucial for interpreting the complex marine environment, and the effective and accurate analysis of these data is particularly important for coastal engineering. In this regard, this study proposes a Physically Informed ViT-GAN (PI-ViT-GAN) automatic partitioning method, based on CFOSAT satellite wave spectrum data. Specifically, the model consists of a generator and discriminator. The generator utilizes a contrastive learning strategy as pretraining and through the self-attention mechanism of the ViT model, it focuses on key parts of the spectrum to extract wave group features and wave element parameters. Partitioning-head joint training realizes the output of wave group partition element indices. Subsequently, the discriminator uses the wave group features and a parametric model for spectrum reconstruction and computes the error with the original observed spectrum to evaluate the partition and reconstruction effects. Additionally, this model incorporates two physically corrected functions, wave system classification loss and merging loss, based on the wave age criterion, thereby guiding the training process, and enhancing model efficiency. The results indicate that the reconstructed theoretical spectrum, obtained through the utilization of this method, aligns well with the original sea wave spectrum, demonstrating a precision superior to the spectral partitioning product of CFOSAT's own SWIM. Combining the robust learning capability of the transformer and the regularization of physical prior knowledge, this model can achieve precise, low-cost automated analysis of satellite wave spectra, providing a new scalable method for big data analysis in marine and coastal engineering.

Внедрение методов анализа больших данных (Big Data) для управления рисками в нефтегазовом комплексе

в данной научной статье рассматривается внедрение инновационных методов анализа больших данных (Big Data) для эффективного управления рисками в нефтегазовом комплексе. Актуальность данной темы обусловлена растущей необходимостью повышения эффективности и безопасности операций в нефтегазовой отрасли, которая сталкивается с множеством рисков, таких как аварии, простои оборудования, колебания цен на энергоносители и геополитическая нестабильность. Целью исследования является разработка комплексного подхода к управлению рисками на основе анализа больших данных, который позволит нефтегазовым компаниям принимать более обоснованные решения и минимизировать потенциальные убытки. В ходе исследования были использованы различные методы, включая сбор и обработку больших объемов структурированных и неструктурированных данных из различных источников, таких как датчики на месторождениях и НПЗ, финансовые отчеты, социальные сети и новостные ленты. Для анализа данных применялись передовые технологии, такие как машинное обучение, глубокое обучение, обработка естественного языка и предиктивная аналитика. Были разработаны специализированные алгоритмы и модели для выявления закономерностей, аномалий и потенциальных угроз в режиме реального времени. Результаты исследования показали, что внедрение методов анализа больших данных позволяет существенно повысить эффективность управления рисками в нефтегазовом комплексе. Например, с помощью предиктивной аналитики удалось снизить количество незапланированных простоев оборудования на 23% и сократить затраты на техническое обслуживание на 18%. Анализ данных о ценах на нефть и газ в сочетании с геополитической информацией позволил оптимизировать торговые стратегии и увеличить прибыль на 12%. Кроме того, выявление потенциальных угроз безопасности, таких как утечки и аварии, с помощью анализа данных с датчиков и камер наблюдения, позволило предотвратить серьезные инциденты и снизить экологические риски. Полученные результаты демонстрируют высокую эффективность применения методов анализа больших данных для управления рисками в нефтегазовом комплексе. Внедрение этих методов позволяет не только минимизировать потенциальные убытки, но и повысить общую эффективность и безопасность операций. Дальнейшие исследования в этой области могут быть направлены на разработку более совершенных алгоритмов и моделей, а также на интеграцию методов анализа больших данных с другими инновационными технологиями, такими как Интернет вещей и блокчейн. this scientific article discusses the introduction of innovative methods of big data analysis (Big Data) for effective risk management in the oil and gas industry. The relevance of this topic is due to the growing need to improve the efficiency and safety of operations in the oil and gas industry, which faces many risks such as accidents, equipment downtime, fluctuations in energy prices and geopolitical instability. The aim of the study is to develop an integrated approach to risk management based on big data analysis, which will allow oil and gas companies to make more informed decisions and minimize potential losses. Various methods were used during the research, including the collection and processing of large amounts of structured and unstructured data from various sources, such as sensors at fields and refineries, financial reports, social networks and news feeds. Advanced technologies such as machine learning, deep learning, natural language processing and predictive analytics were used to analyze the data. Specialized algorithms and models have been developed to identify patterns, anomalies and potential threats in real time. The results of the study showed that the introduction of big data analysis methods can significantly improve the effectiveness of risk management in the oil and gas industry. For example, predictive analytics has reduced the number of unplanned equipment downtime by 23% and reduced maintenance costs by 18%. The analysis of oil and gas price data in combination with geopolitical information allowed us to optimize trading strategies and increase profits by 12%. In addition, the identification of potential security threats, such as leaks and accidents, through the analysis of data from sensors and surveillance cameras, has prevented serious incidents and reduced environmental risks. The results obtained demonstrate the high efficiency of using big data analysis methods for risk management in the oil and gas industry. The implementation of these methods allows not only to minimize potential losses, but also to increase the overall efficiency and safety of operations. Further research in this area may be aimed at developing more advanced algorithms and models, as well as integrating big data analysis methods with other innovative technologies such as the Internet of Things and blockchain.

텍스트마이닝을 활용한 조직핵심인재 역량에 대한 직장인들의 인식

The purpose of this study is to investigate and analyze the competencies of organizational core talents as perceived by employees. To achieve this, interviews and survey questions were created through a review of prior research, using both structured and unstructured questions. Subsequently, data from 134 employees, excluding insincere respondents, were collected through offline interviews and online surveys. The analysis utilized text mining techniques, a big data analysis method, and was conducted by converting descriptive response content into text files using R 4.3.2. The research methodology began with a keyword frequency analysis, followed by visualization through cloud analysis. Then, a keyword difference analysis between organization sizes was conducted using TF-IDF weight comparison, and meaningful word pairs were extracted and categorized through N-gram semantic network analysis, with interview content provided as examples. The results revealed that employees perceive the core competencies of core talents to include problem-solving abilities, critical thinking skills, creativity, and communication skills, recognizing their contribution to project execution and work tasks. Based on these findings, the study discusses its significance, implications, limitations, and directions for future research.

Potential value of expression of receptor accessory protein 4 for evaluating the prognosis of lower-grade glioma patients.

REEP4 is involved in the regulation of the biological process of mitosis. Lower grade glioma (LGG), as a malignant tumor, is accompanied by abnormalities in mitosis, but there have been no reports of REEP4 so far. We collected transcriptome data, DNA methylation data and the clinical characteristics of thousands of patients with LGG. Various big data analysis methods and molecular biology experiments were employed to reveal the impact of REEP4 on the pathological process of LGG. It was found that the expression of REEP4 was significantly elevated and negatively regulated by its methylation site. Therefore, both the high expression of REEP4 and low methylation state of cg16311504 showed that the patients are correlated with lower patient survival rate. In addition, high REEP4 expression participates in the regulation of various cancer-related cellular signaling pathways, such as the cell cycle, MAPK signaling pathway, NOD-like receptor signaling pathway, etc. More importantly, the level of immune cell infiltration significantly increased in the high expression group of REEP4 in the LGG tumor microenvironment and REEP4 has a high positive correlation with PD-L1 and other immune checkpoints. In brief, this study is the first to introduce REEP4 in malignant tumors, which can be used as an independent risk factor that participates in the malignant process of LGG. More importantly, REEP4 has the potential to become a new star in the field of anti-tumor treatment.

A semantic big data analysis method based on enhanced neural networks in IoT

AbstractDue to the growth of neural networks, the semantic big data analysis method can classify images at the pixel level, which is very suitable for the needs of IoT. In semantic big data analysis methods, the DeepLab algorithm is an improved and highly accurate algorithm based on enhanced neural networks. However, the DeepLab algorithm does not fully utilize global information, resulting in poor performance for complex scenes. Therefore, this article makes improvements by introducing a global context information module and providing prior information of complex scenes in images. It extracts global information and merges with original features. It improves the expression ability of features. This global context can enhance the accuracy of semantic big data analysis method, and an attention mechanism is designed. The experimental results display that the improved DeepLab semantic big data analysis method based on self‐attention and global context module has good average pixel accuracy and average intersection to union ratio performance on the Pascal VOC 2012 dataset. And the improvement effect is significant.

Application of Big Data Analysis in Identifying Problems in Financial Reporting of BYD Company

With the advancement of technology, big data has been applied to various industries and fields. Its emergence has also sparked a revolution in accounting information processing and transmission technology. In the era of big data, the environment for corporate financial management is changing. For the new energy vehicle industry, the government's subsidy intensity for related enterprises has been continuously shrinking since 2017. In addition, the limited capabilities of enterprises themselves have brought enormous financial pressure to new energy vehicle enterprises. Therefore, for new energy vehicle enterprises, it is particularly important to avoid financial risks and enhance financial stability in this period. Therefore, this article takes BYD Company as a case study and uses big data analysis methods to identify the problems in cash flow, operational capacity, and debt repayment ability in financial reports, and proposes several improvement measures to adapt to the requirements of enterprise development in the new era of big data.

A big data analytics method for the evaluation of maritime traffic safety using automatic identification system data

The complex traffic situations are among the factors influencing maritime safety. They can be quantitatively estimated through the analysis of traffic data. This paper explores the impact of complex traffic situations on maritime safety, focusing on inland waterway traffic. It presents a big data analytics method, utilizing data from the Automatic Identification System (AIS) and historical maritime accident records. The methodology involves AIS data preprocessing and spatial autocorrelation models, including Moran's index, to extract and evaluate the dynamic characteristics of maritime traffic. The analysis of traffic characteristic includes a thorough investigation into the spatial-temporal distribution of ship average speed and trajectory density. The paper then introduces an effective traffic characteristic analysis model that evaluates the relationship between maritime traffic patterns and accidents. The study, specifically targeting the Nanjing section of the Yangtze River, reveals variations in ship trajectory density and average speed over time. It identifies several hotspots with a significant local correlation between these factors. Moreover, a substantial correlation is found between the locations of maritime accidents and areas with increased ship trajectory density and average speed. These results may provide insights for traffic safety management and highlight strategies for preventing maritime accidents.

Efficiency evaluation of China's transportation system considering carbon emissions: Evidence from big data analytics methods

China's transportation sector is a vital link between production and consumption, but it also has issues with low efficiency, high carbon emissions, and technological bottlenecks. To improve efficiency and provide actionable recommendations and strategies, this study first constructs a comprehensive evaluation index system to gauge the transportation sector's inputs using panel data from different Chinese provinces from 2007 to 2021. Within the assessment system, the principal component analysis (PCA) method is used to reduce the dimension of the indexes, thereby yielding a set of adjusted inputs. Subsequently, the transportation system efficiency (TSE) is evaluated using the super-efficiency SBM-DEA model, which includes unexpected outputs such as carbon emissions, and three-stage DEA modifies the efficiency. Then, we calculate the Malmquist-Luenberger index (TML) and its components: technological change (TTC) and technological efficiency change (TEC). Lastly, the influential factors impacting TSE are analyzed via a truncated regression Tobit model. The following are the conclusions: (1) The transportation industry in China exhibits inefficiency, and the average TSE in Stage I and III is 0.91 and 0.93, respectively. TSE is underestimated due to the influence of external environmental factors and inefficiencies in management in Stage I. (2) TSE in the eastern area also produces significant carbon emissions that surpass the national average. At the same time, other regions face efficiency limitations due to geographical constraints and management obstacles. (3) Insufficient technical capacity is a major cause of inefficiency in the transport sector and is prevalent in the northeast, west, and central regions. (4) Population growth and income per capita advancements foster transportation industry development, while increased GDP, fiscal revenues, and traffic accidents contribute to declining efficiency. The study above findings serve as a foundation for regional and national management initiatives and policies to enhance transportation effectiveness.

Analysis on the Influence of Tax Reduction and Fee Reduction Policies on Digital Development Level of Enterprises Based on AHP-DEMATEL

Since structural tax cuts were first incorporated into active fiscal policy in 2008, various tax breaks have been introduced. Based on the detailed enumeration of tax reduction policy measures, this study constructs an evaluation index system for the level of digital development of enterprises in terms of strategy implementation, operation status, technology level, and quality management. Then DEMATEL hierarchical analysis method and big data analysis method are used to analyse and determine the comprehensive influence degree of each index. Finally, this paper analyses the benefits and convenience brought by tax reduction and fee reduction to the digital development of enterprises and puts forward countermeasures and suggestions for the development of digital transformation of enterprises in the context of tax reduction and fee reduction. This study aims to put forward relevant countermeasures for enterprises to enhance their digital transformation and development capabilities.

빅데이터에 나타난 유아 코딩교육에 대한 사회적 인식 변화

Objectives This study uses semantic network analysis, a big data analysis method, to examine the flow of social awareness and demands for early childhood coding education and identify key keywords to improve understanding of the flow of early childhood coding education and to improve early childhood coding education in the future. The purpose is to find implications for the direction of institutional and policy research.
 Methods For this purpose, the entire period, 2007-2023, was divided into a total of 4 quarters such as from September 25, 2007 to September 25, 2011, September 26, 2011 to September 25, 2015, September 26, 2015 to September 25, 2019, September 26, 2019 to September 25, 2023. Afterwards, data was collected using ‘Early childhood coding education’ as a search term in TEXTOM. Next, we conducted emotional vocabulary analysis of TEXTOM based on the collected data. In addition, after two purification processes, a total of 200 keywords, 50 for each quarter, were extracted based on TF-IDF. Finally, CONCOR analysis was performed with the final keywords.
 Results As a result of the study, social interest in early childhood coding education increased over time, content was expanded into various channels, and positive perceptions were consistently higher than negative. In addition, 50 keywords were extracted for each quarter, and as a result of CONCOR analysis, 4 word groups were formed for each quarter. Through 50 keywords and 4 word groups for each quarter, we were able to identify the social awareness and flow of early childhood coding education by quarter.
 Conclusions This study identified the flow of social awareness and demands for early childhood coding education, and based on this, it can be used as basic data to identify people's major interest in early childhood coding educa-tion and prepare policy measures for the development of early childhood coding education.

Understanding the University Student Experience Through Big Data Analytics

Despite best efforts, the student experience remains poorly understood. One under-explored approach to understanding the student experience is the use of big data analytics. The reported study is a work in progress aimed at exploring the value of big data methods for understanding the student experience. A big data analysis of an open dataset of student comments is being undertaken. The first and simplest use of big data analytics is for the identification of high frequency keyword groups, which, without big data analytics, would be extremely time consuming. However, the lack of context surrounding keyword groups severely limited the ability to draw meaningful conclusions and highlighted the need for human intervention in the analysis process. Future work includes sentiment analysis. This initial work is an impetus for further exploration of big data analytics methods in qualitative contexts, especially in dynamic contexts where rapid data analysis can form a basis for timely interventions.

Big data analysis method based on time series temperature field data of substation transformer

Transformers are one of the main electrical equipment in the power system, and their operating status directly determines the safety, stability, and reliability of the power supply in the power system. The temperature or temperature rise of a transformer is the most important technical parameter that describes the operating status (accidents, faults, and abnormal operation) of the transformer. The following mainly introduces that the monitoring area is gridded by modeling the three-dimensional reality of the transformer, and then the collected infrared temperature data is accurately mapped into the monitoring grid in three dimensions to form the patrol time sequence temperature big data. Through trend analysis, temperature difference and temperature rise analysis, temperature field gradient analysis and other models, the equipment temperature hidden danger detection focusing on prevention is realized.

Intelligent Processing of Power Operation Data Based on Improved Apriori Algorithm

This paper addresses several problems in the power system. Key challenges include low-power information integration, inappropriate system data management, inaccurate system data updating, and inefficient fault diagnosis. We focus on analyzing and diagnosing transmission line faults using the operation data of the power system. The study incorporates the quantitative identification of statements. This is done using the Apriori big data analysis and calculation method. Additionally, we utilize big data analysis and vast power operation data. We aim to achieve automatic analysis and pinpoint the causes of transmission line faults. Furthermore, we seek to optimize the traditional Apriori calculation method. This optimization results in a reduction of about 52% in the candidate item set calculation. The optimized M-Apriori calculation method can analyze the correlation between event index data and faults in real time, and realize automatic diagnosis and analysis of faults through operation data.

Directions for using big data analytics in logistics management

Logistics operations are becoming increasingly complex and require accurate data for effective management. The use of big data in logistics management is a relevant issue due to the growing volume of data and the need to optimize delivery and inventory management processes to meet market demands. The purpose of the study was to develop ways to optimize the management of big data analysis in logistics. To achieve this goal, the methods of analysis, experimentation, and comparison were used. As a result of the study, strategies for optimizing logistics management of big data analysis were developed and successfully applied. The Python programming language based programme effectively optimizes delivery routes using a clustering algorithm and visualizes the results of this process. Additionally, an informative diagram has been drawn up to illustrate the key stages of the developed strategies. The study also developed and presented a table describing the use of big data analysis methods in various logistics companies. The companies were compared in terms of functionality, data, results, and field of activity. It is established that the use of machine learning methods and optimization of data storage and processing significantly increases the efficiency of logistics operations. The results of this study can be used by logistics companies of any size, as well as enterprises engaged in supply chain management. In addition, the recommendations and strategies developed in this study may be useful for information technology and data analytics professionals involved in the development of software solutions and systems to optimize logistics processes

Big Data Analysis Methods for Higher Education Physical Education Teaching Interactive Platform and Its Teaching Effectiveness

Abstract The use of big data comprehensively serves physical education, so that students can clarify their learning situation and make reasonable changes in physical education teaching. In this paper, firstly, we do an analysis and investigation on the application of big data in physical education teaching in colleges and universities, apply the tree hierarchy algorithm to collect the key data indexes of physical education teaching classrooms, and use the SMDA aggregation algorithm to aggregate and analyze the collected key data. The weight of each index system of teaching evaluation is calculated using the hierarchical analysis evaluation method. The role and application of analytical methods in physical education teaching are discussed in the context of big data. This paper constructs a system for evaluating teaching effects, which includes 5 primary indicators and 20 secondary indicators. The evaluation results of the 10-factor index system in the teaching process are divided into five grades to obtain the distribution intervals of excellent, good, medium, qualified, and poor students’ learning performance. The results of data analysis of the attributes of this paper show that the percentages of Cluster l, Cluster 2, and Cluster 3 are 22%, 66%, and 12% respectively. The weights of teaching ability, method, content, attitude, and effect in the actual teaching process were 20.97, 24.29, 24.08, 3.89, and 6.31, respectively. This paper elaborates on the effects and roles of four aspects of the interactive platform Big Data application in physical education teaching and learning in the context of Big Data analytics methods. Ideas for the use of big data analysis methods in sports teaching in colleges and universities are presented.

Big data analysis methods and cultural value mining for the historical research of ceramic art

Big data analysis methods and cultural value mining for the historical research of ceramic art

Identifying Intellectual Structure of Geosciences from the Highly Cited Papers

Understanding the intellectual structure of geosciences is crucial for identifying emerging topics, which can guide future research and funding strategies. This study employs Latent Dirichlet Allocation (LDA) topic modeling and co-word network analysis to explore the intellectual landscape of highly cited geoscientific papers. Through empirical analysis, the study identifies 15 research areas, including climate change, geological processes, environmental impacts, and advancements in data analysis and remote sensing. These findings highlight the prominent role of big data analysis and machine learning methods across various geoscientific domains. Additionally, a notable gap has been identified in the integration of these computational methods with research on organic chemistry and formation processes, suggesting a potential direction for future exploration and innovation.

Research on Big Data Analysis Methods Based on Artificial Intelligence Technology

Research on Big Data Analysis Methods Based on Artificial Intelligence Technology

A big data analytics method for the influence of WeChat public platform articles on blood donation behavior

PurposeThe purpose of this paper is to explore the impact and mechanism of WeChat public platforms articles (abbreviated as WPP) on blood donation behavior using data of WPPA and donation behavior data.Design/methodology/approachThis paper uses multiple linear regression methods, web crawlers and natural language processing technology. It first quantifies the impact of WPP published articles on donation behavior. On this basis, it then selects data from the day of article publication to further study the impact of article dissemination on donation behavior from the perspective of reading quantity, and analyzes the influencing factors of article reading quantity.FindingsThe results show that on the same day that an article is published, there is an increase of 13.8 and 14.3% in blood donation volume and fan registrations, respectively. The mediating effect exists. However, the day after an article is published, there is no longer any effect on blood donations. With a 1% increase in reading quantity, blood donation volume on the day of article publication increases by 0.13%, and this positive impact is promoted by the quality of the articles. A conc ise articles title and body and rich images help drive reading quantity. Moreover, blood donors prefer to read articles about blood dynamics and donation promotion, while articles about news, announcements and administrative affairs make them less inclined to read.Originality/valueFirst, it focuses on WPPA, quantifies the impact of articles on blood donation behavior and analyzes the mechanism. Second, the authors study the impact and timeliness of social media article dissemination to address the insufficiency of existing research. Third, the study provides a scientific basis for the editing and publishing of articles, helping blood banks improve the effectiveness of publicity and recruitment.

Online Communities as a Form of Organizing Narrative Content about SARS-CoV-2 Vaccination: Big Data Analysis

Distrust of vaccines is one of the key problems of modern healthcare both in Russia and abroad. Mass vaccination against SARS-CoV-2 provoked a large number of social media narratives. Online communities play an important role in the formation and dissemination of narratives. The complex system of numerous online platforms for discussing the topic of vaccination on a social network deserves a special description.
 The purpose of the study is to construct a typology of all significant communities of the popular Russian social network “VKontakte” that publish content on the topic of vaccination against COVID-19 (01/01/2020 – 03/01/2023) and describe it in terms of lobbied narratives. Big data collection and analysis methods were used, as well as narrative analysis. Of the 131,119 communities that published at least one vaccination message during the study period, 691 were selected based on self-positioning criteria (name, description) and/or the number of SARS‑CoV-2 vaccination messages above the established threshold. 14 thematic categories of communities have been identified, the self-positioning of which allows reconstructing narratives about forced vaccination and the right to choose; health care and debate about the benefits/effectiveness of vaccination; conspiracy theories; beneficiaries and trust in the authorities; pseudoscience, vaccination myths and scientific knowledge; childcare and parental rights; contrasting medical science with religious beliefs.
 Narratives formed by online communities signify a pronounced polarization of opinions in the digital society regarding SARS-CoV-2 vaccination. The “pro-waxer” and “anti-vaxxer” narratives are reconstructed. Quantitatively “anti-vaxxer” communities significantly predominate over “pro-waxer” ones.