Massive Data Analysis Research Articles

Massive data analyses show negative impact of type 1 and 2 diabetes on the outcome of periodontal treatment

ObjectivesThe aim was to evaluate the impact of diabetes on the outcome of periodontal treatment based on massive data analyses.Materials and methodsData originated from the database of a major German National Health Insurance. Patients who underwent periodontal treatment were allocated to four groups according to their medical condition: type 1 diabetes (D1), type 2 diabetes with the intake of oral anti-diabetics (D2M), type 2 diabetes without the intake of oral anti-diabetics (D2), and a control group without diabetes (ND). Four-year Kaplan-Meier survival analyses on the patient level and multivariate regression analyses were conducted for tooth extraction.ResultsOf 415,718 patients, 4139 matched the criteria for D1, 22,430 for D2M, and 23,576 for D2. At 4 years, the cumulative survival rate (no extraction) was 51.7% in the D1 group, 54.0% in the D2M group, and 57.7% in the D2 group. The ND control group had a significantly higher survival rate of 65.9% (P < 0.0001). In the multivariate analyses, both diabetes types were significantly associated with further tooth loss after periodontal treatment.ConclusionsThe diagnosis of diabetes type 1 or 2 seems to be associated with a higher risk of tooth loss after periodontal treatment.Clinical relevanceThe long-term prognosis of teeth in diabetes patients should be judged carefully.

CheetahVIS

Recently, the spatial-temporal data of urban moving objects, e.g., cars and buses, are collected and widely used in urban trajectory exploratory analysis. Urban bus service is one of the most common public transportation services. Urban bus data analysis plays an important role in smart city applications. For example, data analysts in bus companies use the urban bus data to optimize their bus scheduling plan. Map services providers, e.g., Google map, Ten-cent map, take urban bus data into account to improve their service quality (e.g., broadcast road update instantly). Unlike urban moving cars or pedestrians, urban buses travel on known bus routes. The operating buses form the "bus flows" in a city. Efficient analyzing urban bus flows has many challenges, e.g., how to analyze the dynamics of given bus routes? How to help users to identify traffic flow of interests easily? In this work, we present CheetahVIS, a visual analytical system for efficient massive urban bus data analysis. CheetahVIS builds upon Spark and provides a visual analytical platform for the stakeholders (e.g., city planner, data analysts in bus company) to conduct effective and efficient analytical tasks. In the demonstration, demo visitors will be invited to experience our proposed CheetahVIS system with different urban bus data analytical functions, e.g., bus route analysis, public bus flow overview, multiple region analysis, in a real-world dataset. We also will present a case study, which compares different regions in a city, to demonstrate the effectiveness of CheetahVIS.

Research on Data Processing Architecture of NQI Service Platform in Intelligent Measurement Domain

At present, the quality control and management of smart meters exist in production enterprises and power grid companies respectively. There are some problems, such as weak system, lack of coordination, and lack of overall technical solution ability. By carrying out the design of NQI service platform to support the quality improvement of smart meters, the quality data collection and quality monitoring of smart meters in multiple factors, dimensions and the whole process are realized, and the quality level of smart meters is promoted comprehensively. The NQI service platform focuses on the storage and processing of massive quality data of smart meters, improving the efficiency of quality data mining and utilization, and achieving the overall quality improvement of smart meter industry. In this paper, through the analysis and research of service scenario, the massive quality data analysis and processing mechanism of intelligent power meter is designed.

Big Data Driven Marine Environment Information Forecasting: A Time Series Prediction Network

The continuous development of industry big data technology requires better computing methods to discover the data value. Information forecast, as an important part of data mining technology, has achieved excellent applications in some industries. However, the existing deviation and redundancy in the data collected by the sensors make it difficult for some methods to accurately predict future information. This article proposes a semisupervised prediction model, which exploits the improved unsupervised clustering algorithm to establish the fuzzy partition function, and then utilize the neural network model to build the information prediction function. The main purpose of this article is to effectively solve the time analysis of massive industry data. In the experimental part, we built a data platform on Spark, and used some marine environmental factor datasets and UCI public datasets as analysis objects. Meanwhile, we analyzed the results of the proposed method compared with other traditional methods, and the running performance on the Spark platform. The results show that the proposed method achieved satisfactory prediction effect.

Supervised Sentiment Analysis of Science Topics

Twitter is one of the largest sources of real-time information on the Internet and is continuously fed by millions of users around the world. Each of these users publishes text messages with their opinions, concerns, information, or simply their daily happenings. It is a challenge to address the analysis of massive data in the network, just as it is an objective to look for ways to understand everything that data can offer today in terms of knowledge of society and the market. The sector of science communication is still discovering everything that the web 2.0 and social networks can offer to reach all audiences. This article develops a classification model of messages launched on Twitter, on science topics, in Spanish, with machine learning techniques. The training of this type of models requires the creation of a specific corpus in Spanish for the subject of science, which is one of the most laborious tasks. The classifier is able to predict the sentiment of the message in real time on Twitter, with a confidence interval greater than 80%. The results of its evaluation are at 72% accuracy.

A Novel Method to Protect Private Information in Big Data Environment

In the big data environment, through the analysis and induction of massive data, mining out potential models and studying the market operation rules, can help enterprises adjust strategies, reduce risks and improve efficiency. However, with the wide application of new technology, the sensitive information of users may be disclosed while mining rules. In the big data environment, the amount of data related to users is large. If the privacy information in these data is leaked, it will cause great harm to users. In order to avoid the disclosure of sensitive information in the process of users accessing services in the big data environment, our research will propose new method to protect the privacy data of users in big data environment. Our research design a new approach based on private matching protocol using block chain technology to protect private data in big data environment.

Deep neural network combined with MapReduce for abnormal data mining and detection in cloud storage

Aiming at the abnormal data behavior such as huge amount of data and easy to be stolen or lost in the process of distributed cloud computing in cloud storage environment, an abnormal data mining and detection algorithm of MapReduce based on Hadoop distributed file system (HDFS) and deep neural network is proposed. Firstly, the algorithm analyzes the MAC timestamp characteristics generated by HDFS folder replication, establishes the detection and measurement methods of replication behavior, and ensures that all the patterns that lead to data anomalies, including theft, packet loss and malicious attack, can be detected. Secondly, the algorithm combines deep neural network to design a task partition strategy suitable for arbitrary MapReduce data, and records the input dataset of HDFS hierarchical relationship. Finally, combined with the parallel processing ability of MapReduce, the efficient analysis of massive timestamp data is realized by designing the dataset and algorithm execution scheme suitable for MapReduce task partition. The experimental results show that the algorithm can control the missed detection rate and the number of false detection folders through the segmentation detection strategy. Compared with the existing big data anomaly detection method, the algorithm has higher execution efficiency and good scalability.

Emergence of New Disease: How Can Artificial Intelligence Help?

Emergence of new disease remains a critical parameter in human health and society. Advances in artificial intelligence (AI) allow for rapid processing and analysis of massive and complex data. In this forum article, the recent applications across disease prediction and drug development in relation to the COVID-19 pandemic are reviewed.

Enhanced Generalized Alpha Plane for Numerical Differential Protection Applications

This paper revisits the generalized alpha plane (GAP) formulation applied to differential protection of multi-terminal apparatus. The GAP formulation is improved by aiming for three main goals: 1) control over the internal fault settlement region (FSR); 2) enabling a simple operation characteristic instead of a restraint characteristic; and, 3) The adjustments are not dependent to the device type. For this purpose, the GAP formulation is redefined in terms of two settings, which transform the FSR into a circumference with a controllable center and radius. Thereby, a circular operation characteristic with an adjustable radius is enabled around the FSR, for to improve security. To validate and test the performance of the improved GAP, numerous computer simulations have been carried out using Alternative Transients Program on different multi-terminal equipment (i.e., busbars, transmission lines, and power transformers). Furthermore, massive data analysis is held as a means to highlight the GAP's sensibility. The obtained results demonstrate a well-stable pre-fault point, a consistent circular FSR, and the effectiveness of the improved GAP even under severe conditions.

3D scene and geological modeling using integrated multi-source spatial data: Methodology, challenges, and suggestions

The visualization and analysis of massive, multi-source, multi-dimensional, heterogeneous, dynamic, and complex spatial data is playing an increasingly important role in 3D scene and geological modeling. There are great differences in various research fields regarding data acquisition techniques, morphological descriptions, and application purposes of spatial objects. Therefore, based on an existing three-dimensional visualization technology and analysis program of spatial data, an integrated processing method for the effective fusion and analysis of multi-source complex spatial data is proposed. This method combines a massive point cloud model with a digital elevation model, in addition to combining the complex stratum with fault modeling technology. The structural grid and geological grid are effectively combined to realize the unified modeling and visualization of natural and artificial objects. This method was then applied to a 3D modeling scene of deeply buried mines and a geological construction model for open pit mines. The comprehensive and effective utilization of multi-source geological data is realized by using this modeling method. Finally, the challenges of spatial data visualization from the aspects of the difficulties in spatial data acquisition, the complexities of the spatial structure, the constraint relations of geological bodies, and the complexity of the 3D modeling process are discussed in this study. Three specific suggestions are proposed. This study enhances the application efficiency of multi-source spatial data modeling technologies already in practice and provides reference for future applications and developments.

Research on Data Mining and Investment Recommendation of Individual Users Based on Financial Time Series Analysis

With the continuous development of financial information technology, traditional data mining technology cannot effectively deal with large-scale user data sets, nor is it suitable to actively discover various potential rules from a large number of data and predict future trends. Time series are the specific values of statistical indicators on different time scales. Data sequences arranged in chronological order exist in our lives and scientific research. Financial time series is a special kind of time series, which has the commonness of time series, chaos, non-stationary and non-linear characteristics. Financial time series analysis judges the future trend of change through the analysis of historical time series. Through in-depth analysis of massive financial data, mining its potential valuable information, it can be used for individual or financial institutions in various financial activities, such as investment decision-making, market forecasting, risk management, customer requirement analysis provides scientific evidence.

Rethinking and Restructuring of the Rational Way of Data Sharing in Digital Economy: From the Perspective of Data Typing

The rapid development of digital economy in China has created a miracle of economic growth, at the same time, it also raises many legal problems. The economic rule of law is especially prominent. In essence, it involves the identification and attribution of various types of rights and interests related to data, as well as the justification of protection and the appropriateness of protection. China’s current judiciary and legislation pay enough attention to the core issues in the development of digital economy, data confirmation and protection, which provides a good system and operation environment for promoting the development of digital economy. At present, although the existing system design and practice in China, and related theoretical research results show that the relationship and the implementation between data protection and data sharing can be developed in accordance with the logic of private law that is universally applicable, which includes establishing the ownership of each type of data and practicing the private right logic of “behavior-legal interest” judgment based on tort liability law, intellectual property law and anti-unfair competition law. However, the existing research results and legal practice still show a relatively single and solidified state and mode, which fails to flexibly and effectively adapt to the real needs of mass and diversified data circulation and sharing under digital economy.With the rapid development of Internet platform economy and new business forms, the collection, storage, analysis and use of massive data by platform enterprises have brought challenges to the current legal system of market regulation, including improper collection and excessive collection of user data, and improper or illegal restrictions on the use of data by users and/or third party operators. Therefore, the promotion of data sharing has become the current digital economy in the development process of the urgent response to the practical requirements. To realize data sharing and ensure the synchronization of data protection and data sharing is not only an internal requirement of data operation law, but also an external demand to maintain the high-quality development of digital economy. High level of data protection is the only way to achieve efficient data sharing; at the time of realizing efficient data sharing, the construction of high level of data protection will be continuously stimulated. Therefore, under the guidance of the data sharing principle of “differentiated and equal”, the ownership structure based on “dynamic compatibility rights” is established. Different types of data can carry different subjects’ interest demands, while the same type of data can also carry different subjects’ interest demands. In this way, we can bridge the “high cost, low efficiency” of data sharing brought by the construction of private right model with static exclusive right as the core under the perspective of private law, and even bring the defects of potential anti-competitive risks.

Addressing Long-Standing Questions with Advanced Approaches: The 4th B Chromosome Conference

B chromosomes (Bs) are enigmatic accessory genomic elements extensively characterized in diverse eukaryotes. Since their discovery in the beginning of the 20th century, B chromosomes have been the subject of investigation in laboratories all around the world. As a consequence, scientific meetings have dealt with B chromosomes, including the most specific and important conference in the field, “The B Chromosome Conference.” The 4th B Chromosome Conference (4BCC) took place in Botucatu, Brazil, in 2019 and was an excellent opportunity to discuss the latest developments in the B chromosome research field. B chromosome science has advanced from classical and molecular cytogenetics to genomics and bioinformatics approaches. The recent advances in next-generation sequencing technologies and high-throughput molecular biology protocols have led Bs to be the subject of massive data analysis, thus enabling the investigation of structural and functional issues not considered before. Although extensive progress has been made, questions are still remaining to be answered. The advances in functional studies based on RNA, epigenetics, and gene ontologies open the perspective to a better understanding of the complex biology of B chromosomes.

Analysis of Massive Online Medical Consultation Service Data to Understand Physicians' Economic Return: Observational Data Mining Study.

BackgroundOnline health care consultation has become increasingly popular and is considered a potential solution to health care resource shortages and inefficient resource distribution. However, many online medical consultation platforms are struggling to attract and retain patients who are willing to pay, and health care providers on the platform have the additional challenge of standing out in a crowd of physicians who can provide comparable services.ObjectiveThis study used machine learning (ML) approaches to mine massive service data to (1) identify the important features that are associated with patient payment, as opposed to free trial–only appointments; (2) explore the relative importance of these features; and (3) understand how these features interact, linearly or nonlinearly, in relation to payment.MethodsThe dataset is from the largest China-based online medical consultation platform, which covers 1,582,564 consultation records between patient-physician pairs from 2009 to 2018. ML techniques (ie, hyperparameter tuning, model training, and validation) were applied with four classifiers—logistic regression, decision tree (DT), random forest, and gradient boost—to identify the most important features and their relative importance for predicting paid vs free-only appointments.ResultsAfter applying the ML feature selection procedures, we identified 11 key features on the platform, which are potentially useful to predict payment. For the binary ML classification task (paid vs free services), the 11 features as a whole system achieved very good prediction performance across all four classifiers. DT analysis further identified five distinct subgroups of patients delineated by five top-ranked features: previous offline connection, total dialog, physician response rate, patient privacy concern, and social return. These subgroups interact with the physician differently, resulting in different payment outcomes.ConclusionsThe results show that, compared with features related to physician reputation, service-related features, such as service delivery quality (eg, consultation dialog intensity and physician response rate), patient source (eg, online vs offline returning patients), and patient involvement (eg, provide social returns and reveal previous treatment), appear to contribute more to the patient’s payment decision. Promoting multiple timely responses in patient-provider interactions is essential to encourage payment.

Predictability of drug-induced liver injury by machine learning

BackgroundDrug-induced liver injury (DILI) is a major concern in drug development, as hepatotoxicity may not be apparent at early stages but can lead to life threatening consequences. The ability to predict DILI from in vitro data would be a crucial advantage. In 2018, the Critical Assessment Massive Data Analysis group proposed the CMap Drug Safety challenge focusing on DILI prediction.Methods and resultsThe challenge data included Affymetrix GeneChip expression profiles for the two cancer cell lines MCF7 and PC3 treated with 276 drug compounds and empty vehicles. Binary DILI labeling and a recommended train/test split for the development of predictive classification approaches were also provided. We devised three deep learning architectures for DILI prediction on the challenge data and compared them to random forest and multi-layer perceptron classifiers. On a subset of the data and for some of the models we additionally tested several strategies for balancing the two DILI classes and to identify alternative informative train/test splits. All the models were trained with the MAQC data analysis protocol (DAP), i.e., 10x5 cross-validation over the training set. In all the experiments, the classification performance in both cross-validation and external validation gave Matthews correlation coefficient (MCC) values below 0.2. We observed minimal differences between the two cell lines. Notably, deep learning approaches did not give an advantage on the classification performance.DiscussionWe extensively tested multiple machine learning approaches for the DILI classification task obtaining poor to mediocre performance. The results suggest that the CMap expression data on the two cell lines MCF7 and PC3 are not sufficient for accurate DILI label prediction.ReviewersThis article was reviewed by Maciej Kandula and Paweł P. Labaj.

MAPPING URBAN AREAS USING DENSE TIME SERIES OF LANDSAT IMAGES AND GOOGLE EARTH ENGINE

Abstract. Urban information extraction from satellite based remote sensing data could provide the basic scientific decision-making data for the construction and management of future cities. In particular, long-term satellite based remote sensing such as Landsat observations provides a rich source of data for urban area mapping. Urban area mapping based on the single-temporal Landsat observations is vulnerable to data quality (such as cloud coverage and stripe), and it is difficult to extract urban areas accurately. The composite of dense time series Landsat observations can significantly reduce the effect of data quality on urban area mapping. Multidimensional array is currently effective theory for geographic big data analysis and management, providing a theoretical basis for the composite of dense time series Landsat observations. Google Earth Engine (GEE) not only provides rich satellite based remote sensing data for the composite of dense time series data, but also has powerful massive data analysis capabilities. In the study, we chose Random Forest (RF) algorithm for the urban area extraction owing to its stable performance, high classification accuracy and feature importance evaluation. In this work, the study area is located in the central part of the city of Beijing, China. Our main data source is all Landsat8 OLI images in Beijing (path/row: 123/32) in 2017.Based on the multidimensional array for geographic big data theory and the GEE cloud computing platform, four commonly used reducer methods are selected to composite the annual dense time series Landsat 8 OLI data. After collecting the training samples, RF algorithm was selected for supervised classification, feature importance evaluation and accuracy verification for urban area mapping. The results showed that 1), compared with the single temporal image of Landsat 8 OLI, the quality of annual composite image was improved obviously, especially for urban extraction in cloudy areas; 2) for the evaluation results of feature importance based on RF algorithm, Coastal, Blue, NIR, SWIR1 and SWIR2 bands were the more important characteristic bands, while the Green and Red bands were comparatively less important; 3) the annual composite images obtained by the ee.Reducer.min, ee.Reducer.max, ee.Reducer.mean and ee.Reducer.median methods were classified and accuracy verification was carried out using the verification points. The overall accuracy of the urban area mapping reached 0.805, 0.820, 0.868 and 0.929, respectively. In summary, the ee.Reducer.median method is a suitable method for annual dense time series Landsat image composite, which could improve the data quality, and ensure the difference of features and the higher accuracy of urban area mapping.

Futuros traçados e experiências poéticas: cartografia, performance e vigilância

A partir da noção do capitalismo de vigilância, caracterizado pela coleta massiva e análise de dados, apresentaremos como esse tipo de sistema ultrapassa as discussões da tecnologia, tendo forte configuração política e resultados estéticos que instauram novas formas de visibilidade. Com os, cada vez mais comuns, algoritmos de aprendizado de máquina, e a grande quantidade de informações que esses sistemas podem extrair das mais diversas ações cotidianas, é realizada uma construção de perfil detalhada de cada indivíduo, com objetivos de análises, predições de decisão e exposição mediada ao conteúdo mais adequado. Passa-se à construção de uma subjetividade calcada no futuro, mas atualizada no presente com experiências já previstas e calculadas por um sistema de informação. Numa dualidade de informação e sensação, a experiência dá lugar a predição. Partindo de uma contextualização bibliográfica de teóricos que estudam esses assuntos será estabelecido o cenário de discussão. Em seguida, utilizando-se da pesquisa poética em desenvolvimento pelo autor, será apresentado o projeto “Em busca de um lugar”: uma performance audiovisual que utiliza sistemas comerciais de inteligência artificial como dispositivos disparadores de ações poéticas. Na performance, realiza-se aproximações algorítmicas de lugares geograficamente distantes, a partir de imagens de vigilância feitas pelo artista. Como resultado, são criadas novas cartografias de um espaço que é construído com os mecanismos de vigilância, mas que, de certo modo os subverte, tirando a função e a objetividade esperadas pelo sistema. Desse modo, conclui-se, apresentando a subversão das funcionalidades da lógica algorítmica para a criação poética.

Up-regulation of miR-145 may contribute to repeated implantation failure after IVF-embryo transfer by targeting PAI-1.

Repeated implantation failure (RIF) is a major limiting factor in assisted reproductive technology. As miR-145 (also known as MIR145) is up-regulated in patients with RIF, this study asked, what is the molecular mechanism underlying the affect of miR-145 on embryo implantation in RIF? Ishikawa cells were infected with lentivirus containing miR-145 and miR-145 NC. Massive transcriptome data analyses and bioinformatics analysis were used to search for a potential candidate target of miR-145. The expression of the potential candidate target was detected using quantitative reverse transcription PCR (qRT-PCR) and western blotting in the Ishikawa cells infected with lentivirus containing miR-145 or miR-145 NC. Subsequently, a dual luciferase reporter assay was performed to verify whether the potential candidate target was a novel direct target of miR-145. In addition, expression of PAI-1 (plasminogen activator inhibitor 1, also known as SERPINE1) in endometrial tissue from women with RIF and in control endometrial tissue was examined using qRT-PCR and immunohistochemistry. Based on massive transcriptome data analyses and bioinformatics analysis, PAI-1 was regarded as a potential candidate target of miR-145. miR-145 overexpression was achieved in Ishikawa cells. PAI-1 was confirmed as a direct target of miR-145 by bioinformatic analysis, qRT-PCR, western blotting and dual luciferase reporter assay. Further, results from the clinical sample indicated that at both the mRNA and protein levels, PAI-1 expression was down-regulated in endometrial tissues from women with RIF compared with control group women, and this was negatively related to miR-145 expression. The study results suggests that miR-145 may target and down-regulate PAI-1 expression and influence embryo implantation in women with RIF who are undergoing IVF.

Diverse approaches to predicting drug-induced liver injury using gene-expression profiles

BackgroundDrug-induced liver injury (DILI) is a serious concern during drug development and the treatment of human disease. The ability to accurately predict DILI risk could yield significant improvements in drug attrition rates during drug development, in drug withdrawal rates, and in treatment outcomes. In this paper, we outline our approach to predicting DILI risk using gene-expression data from Build 02 of the Connectivity Map (CMap) as part of the 2018 Critical Assessment of Massive Data Analysis CMap Drug Safety Challenge.ResultsFirst, we used seven classification algorithms independently to predict DILI based on gene-expression values for two cell lines. Similar to what other challenge participants observed, none of these algorithms predicted liver injury on a consistent basis with high accuracy. In an attempt to improve accuracy, we aggregated predictions for six of the algorithms (excluding one that had performed exceptionally poorly) using a soft-voting method. This approach also failed to generalize well to the test set. We investigated alternative approaches—including a multi-sample normalization method, dimensionality-reduction techniques, a class-weighting scheme, and expanding the number of hyperparameter combinations used as inputs to the soft-voting method. We met limited success with each of these solutions.ConclusionsWe conclude that alternative methods and/or datasets will be necessary to effectively predict DILI in patients based on RNA expression levels in cell lines.ReviewersThis article was reviewed by Paweł P Labaj and Aleksandra Gruca (both nominated by David P Kreil).

Improvement of usability in user interfaces for massive data analysis: an empirical study

Big Data challenges the conventional way of analyzing massive data and creates the need to improve the usability of existing user interfaces (UIs) in order to deal with massive amounts of data. How the UIs facilitate the search for information and helps in the end-user’s decision-making depends on developers and designers, who have no guides for producing usable UIs. We have proposed a set of interaction patterns for designing massive data analysis UIs by studying 27 real case studies of massive data analysis. We evaluate if the proposed patterns improve the usability of the massive data analysis UIs in the context of literature search. We conducted two replications of the same controlled experiment, one with 24 undergraduate students experienced in scientific literature search and the other with eight researchers who are experienced in biomedical literature search. The experiment, which was planned as a repeated measures design, compares UIs that have been enhanced with the proposed patterns versus original UIs in terms of three response variables: effectiveness, efficiency, and satisfaction. The outcomes show that the use of interaction patterns in UIs for massive data analysis yields better and more significant effects for the three response variables, enhancing the discovery and visualization of the data. The use of the proposed interaction design patterns improves the usability of the UIs that deal with massive data. The patterns can be considered as guides for helping designers and developers to design usable UIs for massive data analysis web applications.