Big Data Analysis Methods Research Articles

An intelligent big data analytics method for two-dimensional non-residential building energy forecasting

The wide use of sensor and information technologies in buildings resulted in the massive generation of data related to its operation. Thus, there is a need for identifying patterns in data that they may use for the optimal operation of the buildings. The case of non-residential building characteristically provides a big volume of data whose analysis requires computational efficient methods. In this paper, we introduce a new big data analytic method that is applicable to forecasting energy demand in non-residential buildings. The goal is to make energy forecasts in a two-dimensional (2D) space defined by i) the electricity load and ii) gas demand. The proposed method combines the matrix profile (MP) method with a Long-Short Term Memory (LSTM) neural network. The combination of the above tools provides an efficient method in 2D hour ahead forecasting with the big data environment of smart buildings. In specific, with respect to mean average percentage error (MAPE) the combined MP-LSTM method provides a concurrent forecast of electricity and gas around 3% and 4%, respectively.

Research on a Regional Landslide Early-Warning Model Based on Machine Learning—A Case Study of Fujian Province, China

China’s landslide disasters are serious, and regional landslide disaster early-warning is one of the important means of disaster prevention and mitigation. The traditional regional landslide disaster early-warning model, however, is limited by the complex landslide induction mechanism, limited data accumulation, and insufficient big data analysis methods, and has problems such as limited early-warning accuracy and insufficient refinement. In this paper, a machine learning method was introduced into the field of regional landslide disaster warning. From the model construction process of training sample-set construction, sample learning and training, model parameter optimization, model preservation, warning output, and so on, a method for constructing a regional landslide early-warning model based on machine learning was systematically proposed. In the sample learning and training, 80% of the training sample-set was used as the training set, and 20% was used as the test set for five-fold cross validation. The Bayesian Optimization algorithm was used to optimize the model parameters, and the accuracy, ROC curve, and AUC value were used to verify the model accuracy and model generalization ability. With China’s Fujian province as an example, based on nine years of geological and meteorological data (2010–2018), geological environment factors, factors of hazard-affected bodies and historical disaster situations, and rainfall-induced factors in four categories, a total of 26 indicators were used as input characteristic parameters. Six machine learning algorithms were adopted to improve model training; the results showed that the Random Forest algorithm performed the best, giving an accuracy of 92.3%, and was the model with the best generalization ability (AUC was 0.955). The second best was the Artificial Neural Network model, with an accuracy of 0.937 and an AUC of 0.935. Next were the Nearest Neighbor model, the Logistic Regression model, and the Support Vector Machine; the poorest results were from the Decision Tree model. Finally, the typical rainfall-type landslide disaster process in Fujian Province was selected as an example to verify the Random Forest algorithm model. The results showed that compared with the early-warning results of the original explicit statistical model, the hit rate of the new model was 6 times, or equal to that of the original model, and the landslide density in the early-warning area of the new model was 1.6–1.7 times that of the original model. Preliminary verification showed that the new model based on the Random Forest method has obvious advantages, a higher hit rate and a smaller warning area, and can achieve more accurate warnings. The follow-up will continue to track the new landslide disaster situation in the study area and carry out model verification and correction.

Comparison of Bigdata Application in User Behavior for Short Video Industry: TikTok and Xiaohongshu

With the progression of innovation, the applications created based on enormous information will gotten to be increasingly copious. Since the computing control of computers and the gigantic sum of information accessible are not easy issue, cleverly applications spoken to by manufactured insights and profound learning will ended up more astute. Within the setting of such industry competition, each brief video stage still must deeply understand the contrasts of its clients in numerous division measurements, so as to realize more precise advertise situating and define more sensible improvement methodologies. This article uses two platforms which analyze user behaviors depending on bigdata. AARRR Model is a technological tool that helps firms to find their problems. Bigdata is a basic section to build up such model. According to the analysis, the two companies built their own big data analysis methods based on this model to help them achieve greater success. These results shed light on guiding further exploration of big data analysis implementation for short video industry.

Simulation modeling of gas exchange processes based on physical factors and effects with the implementation that uses big data analysis methods

The paper presents models that allow calculating and optimizing decompression parameters based on the physical description of gas exchange processes. The main feature of the work is the correction of the known tabular data not only based on the user’s decision, but also using physical models that describe a gas exchange in the body. This can significantly improve the quality of decompression planning. The result is achieved since the difference in decompression processes is primarily described not by physiological, but by physical factors. The work presents a detailed diagram of the modules included in it, describes the principles of their functioning. The following modules are described: a heat loss module, a gas exchange module, a module for optimization of decompression profiles, a microbubble corrections module, a decompression prediction module that calculates decompression risks based on a distributed analysis and data processing. The decompression risk prediction is based on the calculation of the critical safety level, for which a distributed analysis and data processing technique has been created. The implementation of this technique allows forming a detailed idea of the possibilities of its use in other subject areas.

동시출현 단어 분석을 활용한 한⋅중 언어의 대조 분석 연구 동향 탐색

We attempted a Contrastive Analysis(CA) of Korean-Chinese language study using the big data analysis method. The target of analysis is 1,316 research papers in Korean-Chinese language contrast, using the ‘Co-Occurrence Words Analysis’ and ‘Word Network Analysis’ methods. We divided into three periods and looked at the trends of control studies for the ‘Title’ and ‘Abstract’ of paper.BR 1) The CA between Korean and Chinese language has been continuously carried out since 1989, and the analysis contents are gradually becoming more detailed. In addition, research on ‘culture’ of the Korean-Chinese language has been continuously shown since 2010.BR 2) Since 2015, the contrast study of Korean-Chinese language has not been connected to L2 Education, and has remained in the CA study. We hope that the results of CA studies will be more actively linked to L2 Education. And we look forward to starting a CA study that can help ‘L2 Korean-Chinese language education’ from now on.BR 3) In L2 Education research, from the perspective of ‘learners’ and ‘teachers’, most of the CA studies so far have been conducted focusing on ‘learners’. We look forward to more research for ‘teachers’.BR 4) In L2 education research, among ‘difference’ and ‘similarity’, most of the studies so far have been conducted focusing on ‘difference’. We look forward to the start of research analyzing ‘similarity’.

코로나19 펜데믹 전후로 홈트레이닝 트렌드는 어떻게 변화했는가?

Purpose: The purpose of this study is to examine changes in keywords related to ‘home training’ before and after the COVID-19 pandemic using big data analysis methods and to analyze trend changes in detail.BR Method: As the keyword for data search, keyword results for home training in 2018-2019 and 2019-2020, and the top 30 keywords were analyzed to compensate for het limitations of network simplification and network visualization centering on important nodes in the entire network. The UcINet 6 & NetDraw analysis and word cloud visualization were conducted in stages.BR Results: First, as a result of analyzing the top 10 keywords related to home training in the past 2 years from 2018 to 2019, home training, exercise, diet, stretching, body care, YouTube, exercise method, yoga mat, women, and dumbbells were found in that order. Second, as a result of analyzing the top 10 keywords related to home training in the last 2 years from 2019 to2020, home training, Corona 19, exercise, non-face-to-face, VOD, pandemic, application, smart home, social distancing, and smart home were in that order.BR Conclusion: It has confirmed that the trend of home training before and after Corona 19 is changing, and as the origin of the COVID-19 pandemic is over, home training is starting to become a full-fledged fashion, establishing itself as a new exercise trend.

Smart Homes and Families to Enable Sustainable Societies: A Data-Driven Approach for Multi-Perspective Parameter Discovery Using BERT Modelling

Homes are the building block of cities and societies and therefore smart homes are critical to establishing smart living and are expected to play a key role in enabling smart, sustainable cities and societies. The current literature on smart homes has mainly focused on developing smart functions for homes such as security and ambiance management. Homes are composed of families and are inherently complex phenomena underlined by humans and their relationships with each other, subject to individual, intragroup, intergroup, and intercommunity goals. There is a clear need to understand, define, consolidate existing research, and actualize the overarching roles of smart homes, and the roles of smart homes that will serve the needs of future smart cities and societies. This paper introduces our data-driven parameter discovery methodology and uses it to provide, for the first time, an extensive, fairly comprehensive, analysis of the families and homes landscape seen through the eyes of academics and the public, using over a hundred thousand research papers and nearly a million tweets. We developed a methodology using deep learning, natural language processing (NLP), and big data analytics methods (BERT and other machine learning methods) and applied it to automatically discover parameters that capture a comprehensive knowledge and design space of smart families and homes comprising social, political, economic, environmental, and other dimensions. The 66 discovered parameters and the knowledge space comprising 100 s of dimensions are explained by reviewing and referencing over 300 articles from the academic literature and tweets. The knowledge and parameters discovered in this paper can be used to develop a holistic understanding of matters related to families and homes facilitating the development of better, community-specific policies, technologies, solutions, and industries for families and homes, leading to strengthening families and homes, and in turn, empowering sustainable societies across the globe.

AI-based big data analytics model for medical applications

Today, people's daily lives depended on their mobile phones, facilitating many beneficial tasks. Mobile computing and cloud computing combine cloud-based mobile computing to improve their features and overcoming their disadvantages such as memory space, processor, and longer battery life. By combining the four "V's" of volume, variety, velocity and veracity, Big Data analytics tools allow the utility to be extracted through knowledge. This paper explores the development of interconnected healthcare systems and the contribution of mobile cloud applications and big data analysis. With the implementation of cloud computing in the healthcare system, inspiration and development of interconnected health - software and care services are given. Big Data services for the healthcare sector on a cloudlet-based mobile cloud computing platform. Big data analysis methods, equipment, and solutions are examined. We have concluded the development of interconnected medical systems using Big Data and mobile cloud computing.

Deep Learning-Based Intelligent Analysis of Medical Big Data

Data-driven medical big data analysis methods have arisen as the times have demanded, assisting in the intelligent identification of medical health. With the widespread use of computer technology, medical health data has also expanded considerably. It is still challenging to evaluate medical big data, nevertheless because of the heterogeneous format of the data, the large number of missing records and the amount of noise. While deep learning constructs a hierarchical model by imitating the human brain, traditional machine learning techniques are unable to efficiently harvest the rich information included in medical big data. It boasts strong automated feature extraction, intricate model building and effective feature expression among other significant features. It is a deep learning technique that takes features from the original medical imaging data at every level, starting at the lowest. For the purpose of intelligent illness identification and diagnosis, this research builds a deep learning-based data analysis model for medical pictures and transcripts. The model chooses and optimises model parameters using a vast amount of medical big data. It also automatically picks up on the pathological analysis process used by doctors or medical researchers. Finally, the model conducts disease judgement and effective decision-making based on the analysis findings of the medical big data. The outcome of the trial demonstrate that the approach can evaluate large amounts of medical data and achieve early illness detection. In addition, it may evaluate the patient's physical health state based on their physical examination data and forecast their future risk of contracting a certain disease. Greatly lower the workload for physicians or researchers in the medical field and increase the effectiveness of their effort.

Journal-related Activities and Other Special Activities at the 2022 American Society of Anesthesiologists Meeting

Journal-related Activities and Other Special Activities at the 2022 American Society of Anesthesiologists Meeting

Modeling and Analysis Method of National Fitness Big Data for Basketball Projects Based on a Multivariate Statistical Model

How to start from the fitness needs of people and effectively improve the precision of the supply of public fitness services everyone is an important issue that needs to be solved first at the current stage. This requires us to proceed from the reality, conduct accurate research, and find a method that can match the current problem. In this paper, taking basketball projects in national fitness as an example, by introducing a proposition about the development of small basketball events, the corresponding big data modeling and analysis methods are studied. The research methods and research objectives involved in this paper are based on the relevant parameters of the multivariate statistical model. First, the article introduces the calculation principle of the multiple linear regression model. We introduce the concept of variance inflation factor involved in this principle and carry out the modeling and analysis of big data based on this variable. In order to illustrate the application effect of big data in this kind of research, this paper introduces three different big data technologies, including immune selection optimization algorithm, particle swarm optimization algorithm, and Elman neural network, to predict and analyze the variance inflation factor (VIF) corresponding to the small basketball project. The analysis results show that the Elman network exhibits certain advantages in terms of computing convergence time. And, as the number of calculation steps increases, the superiority of the Elman network is more obvious. As far as the prediction performance is concerned, the square of the correlation coefficient corresponding to the immune selection optimization algorithm is the largest and the sum of the squares of the residuals is the smallest, showing superior prediction performance.

Research on Medical Big Data Analysis and Disease Prediction Method Based on Artificial Intelligence.

In recent years, the continuous development of big data, cloud services, Internet+, artificial intelligence, and other technologies has accelerated the improvement of data communication services in the traditional pharmaceutical industry. It plays a leading role in the development of my country's pharmaceutical industry, deepening the reform of the health system, improving the efficiency and quality of medical services, and developing new technologies. In this context, we make the following research and draw the following conclusions: (1) the scale of my country's medical big data market is constantly increasing, and the global medical big data market is also increasing. Compared with the global medical big data market, China's medical big data has grown at a faster rate. From the initial 10.33% in 2015, the proportion has reached 38.7% after 7 years, and the proportion has increased by 28.37%. (2) Generally speaking, urine is mainly slightly acidic, that is, the pH is around 6.0, the normal range is 5.0 to 7.0, and there are also neutral or slightly alkaline. 8 and 7.5 are generally people with some physical problems. In recent years, the pharmaceutical industry has continuously developed technologies such as big data, cloud computing, Internet+, and artificial intelligence by improving data transmission services. As an important strategic resource of the country, the generation of great medical skills and great information is of great significance to the development of my country's pharmaceutical industry and the deepening of the reform of the national medical system. Improve the efficiency and level of medical services, and establish forms and services. Accelerate economic growth. In this sense, we set out to explore.

Analysis of the optimal shooting angle in football matches based on network data mining

As an important research field in video understanding, human behavior recognition is widely used in intelligent monitoring, intelligent nursing, interactive behavior, and robot control. ObjectiveThe purpose of this study is to explore the analysis method of interaction behavior of soccer video. MethodsIn this study, we use scene simulation, big data analysis, experimental comparison, and other methods to analyze the interaction between players in a football video. The resultsof experiment and analysis of attack mode under interactive behavior show that the accuracy of attack direction determination results in traditional video analysis method and accurate analysis method is 75.60 % and 85.2 %, respectively. In this study, taking into account the two factors of staff proportion and average distance, the accuracy is improved to a certain extent, which can reach 91.5 %. It can be concluded that the deviation rate proposed in this study can better describe the position information of the football, can realize the detection of the basic attack mode, and can enable the computer to identify the attack direction of both players in the shortest time, which shows that the method is efficient and accurate. It provides a good analysis method for the research of human interaction behavior in football video and even in all kinds of videos.

The Methods of Deep Learning and Big Data Analysis in Promoting Sustainable Architecture

These days, sustainability in different aspects has been among the main discussions of architecture and building science. At the same time, historic architecture has evolved over centuries and has adapted to environmental conditions, it can be a great source of inspiration in using smart ways to achieve sustainable architecture. A good illustration of this adaptation can be found in using vernacular materials, the spatial configuration according to climate conditions, and different elements of historic architecture that have helped to improve the occupant’s comfort. In response, one plausible solution for improving the sustainability of architecture is translating the concept of the sustainable elements and features of historic architecture to be used in contemporary architecture. Therefore, these elements need to be studied thoroughly to comprehend their features and characters. There are several studies, investigating sustainable historic architecture to find and measure sustainable solutions by using conventional methods. Although the accuracy of studying the sustainable historic elements has been fairly high, the number of features and variety of these elements in historic architecture have made this task highly challenging. It has been suggested to study and evaluate a considerable number of these elements in different historic architecture to reduce the errors and increase the reliability of results. Since the conventional methods are labor-intensive, time-consuming, and costly, this paper proposed a robust AI method to study the sustainable elements of historic architecture by using Deep Learning. In this study, by introducing and developing a new method for detecting sustainable elements in historic architecture, their features were comprehensively extracted by means of mining meaningful data from areal images of historic cities to produce big data. The proposed method has a sophisticated workflow starting from subdividing the High-Resolution Aerial Images to detecting the sustainable elements and using data science to analyze the extracted features of the segmented objects. Results of a sample analysis of this method showed its high accuracy and its applicability in analyzing sustainable elements of historic architecture, by which designers are expected to design more sustainable buildings inspired by historic architecture.

Breast Cancer Cell Data Analysis and Visualization Using R

Breast cancer is the most frequently occurring cancer in women. If the cancer is diagnosed and treated at an early stage, the patient has a survival rate of 99% after 5 years, but it significantly drops to 29% when it reaches a distant stage. Thus, it is very important to detect ‘positive cancer cells’ in the early stage, so I analyzed the 569 breast cancer cell data provided by the University of Wisconsin using R-Studio. Through this program, I visualized the relationships between 10 different cell characteristics, and researched the kind of relationship between radius and positive cancer cells by utilizing graphs. Also, I created a predictive model that can detect positive cancer cells based on logistic regression and training original 569 data. Finally, I recommended the adequate age (35-39) for the breast cancer examination by analyzing the breast cancer statistics. Through this research, I derived effective breast cancer predicting model and deeply explored big data analysis method and data-mining on R-Studio.

A selective review of statistical methods using calibration information from similar studies

In the era of big data, divide-and-conquer, parallel, and distributed inference methods have become increasingly popular. How to effectively use the calibration information from each machine in parallel computation has become a challenging task for statisticians and computer scientists. Many newly developed methods have roots in traditional statistical approaches that make use of calibration information. In this paper, we first review some classical statistical methods for using calibration information, including simple meta-analysis methods, parametric likelihood, empirical likelihood, and the generalized method of moments. We further investigate how these methods incorporate summarized or auxiliary information from previous studies, related studies, or populations. We find that the methods based on summarized data usually have little or nearly no efficiency loss compared with the corresponding methods based on all-individual data. Finally, we review some recently developed big data analysis methods including communication-efficient distributed approaches, renewal estimation, and incremental inference as examples of the latest developments in methods using calibration information.

Automate Short Cyclic Well Job Candidacy Using Artificial Neural Networks–Enabled Lean Six Sigma Approach: A Case Study in Oil and Gas Company

Artificial Neural Networks (ANNs) are a part of Artificial Intelligence (AI) that is commonly used for pattern recognition, regression and classification. This technology allows us to learn historical data and generate patterns from the precedent data. In oil and gas companies, large amounts of data are produced every day. Many accurate decisions in this type of company are made from the data. Cilon Indonesia (CI) Co. Ltd. is one of the oil and gas companies currently operating the largest oil field in Indonesia. This type of company's operation and financial profit depends on oil price, which is affected by global oil supply and demand. If oil prices fall suddenly, all oil and gas companies need to run their businesses more efficiently and effectively. There are many ways to make this kind of company run their business effectively and efficiently by implementing several strategies such as capital cost efficiency, operational cost efficiency and even laying off some employees. One of the major costs in operation in oil and gas companies is the cost for well workover. This well workover does not always produce oil gain. In fact, even it is resulting in oil gain, but not all well workover programs are economical whenever the oil price is low. This condition makes Petroleum Engineer (PE) need to select the best well workover for certain wells. Well candidates for workover are usually selected manually using data from many resources, reports and information. Well candidates are reviewed one by one, and with several criteria, the well is proposed to a certain type of well workover. This research explains how this company improves their selection of well candidates for the most economic workover called Short Cyclic Steam Stimulation (SCSS). The process improvement is done using the hybrid method: lean six sigma method and big data analytics method, which utilize ANNs to predict the oil after workover executed. The result demonstrates how this hybrid method can improve the process with a sustainable solution. Its successful improvement in PE time selects SCSS well candidates from 2 hours to 10 minutes to generate 20 wells per day. Its also improve the success rate of SCSS workover from 61% to 73%.

Application of K-means Clustering Algorithm in Rice Production of Tamil Nadu, India

Aim: (i) To Cluster the Rice data using K-Means clustering algorithm. (ii) To helps the study of crop yield prediction.
 Study Design: K-Means clustering technique is one of the most common exploratory data analysis used to get an intuition about the structure of the data.
 Place and Duration of Study: Time Series crop data were collected from the season and crop report, Directorate of Economics and Statistics, Chennai for the period of 2015-2020.
 Methodology: The machine learning algorithm of big data analytics method such as K-means clustering algorithm helps to predict the paddy yield accurately in Tamil Nadu. The performance of the technique is examined through the determinable value of k by Elbow method and Silhouette method which helps in the crop yield prediction.
 Results: The observed results show that there is a positive relationship between area, production, area under irrigation, minimum temperature, and relative humidity and a close negative relationship with moisture and wind speed. Additionally, two clusters were identified with cluster 2 having the highest mean value, followed by 1. The identification of the highest mean clusters will guide farmers on where best to concentrate on when planting their crops in ordered to improve productivity and crop yield.
 Conclusion: This study reveals a scalable, simple, and reduced method for correctly assessing rice production over a large area using publicly released multi-source data, which may have been used to calculate crop production in areas with rarely observed data and all around the world.

Intelligent Make Policy Support Systemic for Business Management Based on Big Data Analysis

With the development of message technology, big data is connected with business management. The research content and present situation of intelligent make policy support systemic in big data are introduced. Based on the knowledge-based systemic and its relationship with make policy-making process, the knowledge and knowledge processing systemic in IDSS are discussed. With the development of computer and its network, many enterprises have established their own message management systemics. How to observe the enterprise itself and the outside world in a flexible, real-time, vertical and horizontal, dynamic and slicing way, so as to capture relevant message for the development needs of the enterprise. The rapid development of big data has aroused widespread concern and attention at home and abroad. Scientific and effective analysis and processing of big data are the core issues in the field of big data. This paper summarizes the research status of big data analysis at home and abroad from three aspects: analysis as a service, big data analysis methods, and big data-driven science germination through literature review. By analyzing the statistical and semantic features of the existing data, we can find out the rules and then summarize them into an abstract data analysis model, thus providing the basis for data analysis, and the establishment of data graph and model can better reflect the specific content of the study. In the research of big data, it is found that both traditional data and big data have packet loss rates. According to the research, the packet loss rate of traditional data is as high as 70.9%, but that of big data is only 34.5%. Making a formula in the calculation of big data can help us to observe more. Through this research, it is found that big data has a more direct relationship with business management, and it controls various ways of enterprises. The era of big data has come, and all walks of life are faced with unprecedented data volume and data analysis requirements.

Predicting Student Learning Effectiveness in Higher Education Based on Big Data Analysis

With the global higher education entering the era of “quality is king,” the perception and acquisition of learning experience and the evaluation of learning quality and its effect have attracted more and more attention. However, at present, most of the academic evaluations of university courses are based on final examinations, supplemented by appropriate amount of regular tests, and little attention is paid to the development and quality of students’ learning ability in the learning process, which inevitably causes academic evaluations to be equivalent to assessments and deviates from the original purpose of academic evaluations to promote continuous improvement of teaching and learning quality. On this basis, this study uses the big data analysis method to predict the teaching effect of college students, in order to improve the existing teaching problems, grasp the teaching essence, and construct a relatively perfect curriculum evaluation system combined with the course of “teaching effect guidance + teaching action,” so as to further improve the academic evaluation system and improve the teaching quality. This will provide a reference for further improving the academic evaluation system and improving the teaching quality.