Combining Data Mining Research Articles

Combined cycle gas turbine power output prediction and data mining with optimized data matching algorithm

Electrical power output (PE) for a combined cycle gas turbine (CCGT) consisting of 9568 data records collected over a 6-year period is evaluated by the transparent open box (TOB) machine-learning method to provide accurate PE predictions and insight to prediction errors. The PE predictions derived by applying the TOB optimized data matching technique are more accurate than published predictions for the dataset from fifteen correlation-based, machine-learning algorithms. TOB achieves this high-accuracy using a tuning subset of < 150 (~ 1.5%) data records. Its accuracy is confirmed by testing the optimized solutions against all dataset records in 15 runs spread across five shuffled datasets. The dataset has a few extreme outliers associated with its four independent variables and these negatively impact the prediction accuracy of machine-learning methods. Through its transparency and forensic-like auditability of its calculations for individual data records, the TOB algorithm is able to mine the dataset to provide useful insight to the interactions of the outliers with other data records. This enables a filtered dataset (9533 records), excluding 35 carefully selected data records, to be customized to deliver much improved prediction accuracy (RMSE = 2.89 MW). Mining the dataset also reveals significant differences in prediction accuracy achieved for different sectors of the PE distribution. This insight identifies that prediction accuracy could be further improved by dividing the dataset into separately optimized subsets, three along its main PE trend plus a fourth, small subset consisting of the outliers. The TOB algorithm demonstrates its value as a machine-learning tool capable of generating accurate predictions and easily auditable data mining. It is well suited for CCGT efficiency and performance optimization.

Key Technologies of Intelligent Transportation Based on Image Recognition and Optimization Control

With the development of digital image processing technology, the application scope of image recognition is more and more wide, involving all aspects of life. In particular, the rapid development of urbanization and the popularization and application of automobiles in recent years have led to a sharp increase in traffic problems in various countries, resulting in intelligent transportation technology based on image processing optimization control becoming an important research field of intelligent systems. Aiming at the application demand analysis of intelligent transportation system, this paper designs a set of high-definition bayonet systems for intelligent transportation. It combines data mining technology and distributed parallel Hadoop technology to design the architecture and analysis of intelligent traffic operation state data analysis. The mining algorithm suitable for the system proves the feasibility of the intelligent traffic operation state data analysis system with the actual traffic big data experiment, and aims to provide decision-making opinions for the traffic state. Using the deployed Hadoop server cluster and the AdaBoost algorithm of the improved MapReduce programming model, the example runs large traffic data, performs traffic analysis and speed–overspeed analysis, and extracts information conducive to traffic control. It proves the feasibility and effectiveness of using Hadoop platform to mine massive traffic information.

Application of clustering algorithm in social network user scenario prediction

In order to improve the ability of social network user behavior analysis and scenario pattern prediction, optimize social network construction, combine data mining and behavior analysis methods to perform social network user characteristic analysis and user scenario pattern optimization mining, and discover social network user behavior characteristics. Design multimedia content recommendation algorithms in multimedia social networks based on user behavior patterns. The current existing recommendation systems do not know how much the user likes the currently viewed content before the user scores the content or performs other operations, and the user’s preference may change at any time according to the user’s environment and the user’s identity, Usually in multimedia social networks, users have their own grading habits, or users’ ratings may be casual. Cluster-based algorithm, as an application of cluster analysis, based on clustering, the algorithm can predict the next position of the user. Because the algorithm has a “cold start”, it is suitable for new users without trajectories. You can also make predictions. In addition, the algorithm also considers the user’s feedback information, and constructs a scoring system, which can optimize the results of location prediction through iteration. The simulation results show that the accuracy of social network user scenario prediction using this method is higher, the accuracy of feature registration of social network user scenario mode is improved, and the real-time performance of algorithm processing is better.

Systematically Exploring the Antitumor Mechanisms of Core Chinese Herbs on Hepatocellular Carcinoma: A Computational Study.

Objective Chinese herbs play a positive role in the management of hepatocellular carcinoma (HCC) in China. However, it is not clear which of Chinese herbs are critical for the treatment of HCC. Besides, mechanisms of CCHs in the treatment of HCC remain unclear. Hence, our goal is to identify the core Chinese herbs (CCHs) for treating HCC and explore their antitumor mechanism. Methods Firstly, clinical traditional Chinese medicine (TCM) prescriptions for HCC were collected from Chinese National Knowledge Infrastructure (CNKI) database, and then, data mining software was used to identify CCHs. After that, bioactive compounds and corresponding target genes of CCHs were obtained using three TCM databases, and target genes of HCC were acquired from MalaCards and OMIM. Subsequently, common target genes of CCHs and HCC were screened. Moreover, biological functions and pathways were analyzed, and Cytoscape plugin cytoHubba was used to identify hub genes. Finally, prognostic values of hub genes were verified by survival analysis, and the molecular docking approach was utilized to validate the interactions between targets and bioactive compounds of CCHs. Results Eight CCHs were determined from 630 prescriptions, and 100 bioactive compounds (e.g., quercetin and luteolin) and 126 common target genes were screened. Furthermore, common target genes of CCHs and HCC were mainly enriched in cancer-associated pathways, and six hub genes with statistical significance in survival analysis were selected as key target genes for molecular docking. Additionally, molecular docking showed that the bioactive compounds docked well with the protein receptors of key target genes. Conclusion By combining data mining, network pharmacology, molecular docking, and survival analysis methods, we found that CCHs may play a therapeutic role in HCC through regulating the target genes and pathways related to cancer occurrence and development, angiogenesis, metastasis, and prognosis.

Outlier data mining model for sports data analysis

The results of data mining can be used to predict the physical health status of sports athletes and college sports students and provide physical fitness warnings, so that students can pay attention to physical health status and adjust their physical exercise status. Discrete Morse theory, as a powerful optimization theory, plays a big role in algorithm optimization. This paper combines data mining and discrete Morse theory to propose a grid clustering algorithm based on discrete Morse theory. Moreover, according to the theorem that the cell complex reaches the optimum when it has the smallest possible critical point, this study applies the concept of critical points in the discrete Morse theory to optimize the grid clustering process to obtain clustering results. In addition, this study uses the improved C4.5 algorithm to analyze the physical fitness assessment results and obtains a valuable analysis of the physical fitness assessment results.

Image Pattern Recognition Combined With Data Mining for Diagnosis and Detection of Myocardial Infarction

How to apply data analysis algorithms to China's primary hospitals is still a problem that needs to be solved. In order to effectively explore the pathogenesis of myocardial infarction disease, this study collected a large amount of real data as a basis for data analysis through data survey, improved traditional cluster analysis and data mining methods, and proposed effective data mining methods for myocardial infarction. In addition, this study analyzes data sources by implementing clustering analysis algorithms and combines data mining algorithms to provide decision support information for disease research. Finally, this study uses experimental methods of image data analysis combined with data mining methods to record the results. The research shows that the algorithm of this study has certain feasibility and can provide theoretical reference for subsequent related research.

A Combined Approach for Customer Profiling in Video on Demand Services Using Clustering and Association Rule Mining

The purpose of this paper is to propose a combined data mining approach for analyzing and profiling customers in video on demand (VoD) services. The proposed approach integrates clustering and association rule mining. For customer segmentation, the LRFMP model is employed alongside the k-means and Apriori algorithms to generate association rules between the identified customer groups and content genres. The applicability of the proposed approach is demonstrated on real-world data obtained from an Internet protocol television (IPTV) operator. In this way, four main customer groups are identified: “high consuming-valuable subscribers”,” less consuming subscribers”,” less consuming-loyal subscribers” and “disloyal subscribers”. In detail, for each group of customers, a different marketing strategy or action is proposed, mainly campaigns, special-day promotions, discounted materials, offering favorite content, etc. Further, genres preferred by these customer segments are extracted using the Apriori algorithm. The results obtained from this case study also show that the proposed approach provides an efficient tool to form different customer segments with specific content rental characteristics, and to generate useful association rules for these distinct groups. The proposed combined approach in this research would be beneficial for IPTV service providers to implement effective CRM and customer-based marketing strategies.

Semi supervised data mining model for the prognosis of pre-diabetic conditions in type 2 Diabetes Mellitus.

Diabetic Mellitus is the leading disease in the world irrespective of age and geographical location. It is estimated that 43% of the overall population is affected by the disease. The reasons for the disease include inappropriate diet lifestyle with allied symptoms like obesity. Therefore, the prognosis and diagnosis of the disease are important for adequate combat and care. The prognosis related known symptoms of the disease include incontinence (inability to control urination) and frequent fatigue. Moreover, early prediction of the disease plays an important role in the prognosis of other associated conditions such as heart failure leading to chronic illness. Hence, it is of interest to describe a data mining based prediction model using known features (derived from epidemiological data collected from the public hospital using routine tests) to help in the prognosis of the disease. We used data pre-processing techniques for handling missing values and dimensionality reduction models to improve data quality. The Minimum Description Length principle (MDL) model for discretization (replacing a continuum with a finite set of points) is used to reduce high-level dimensionality of the dataset, which enabled to categorize the dataset into small groups in ordered intervals. Thus, we describe a semi-supervised learning technique (identifies promising attributes using clustering and classification methods) by combining data mining techniques for reasonable accuracy having adequate sensitivity and specificity for further discussion, cross-validation, revaluation, and application. Early prediction of the disease with improved accuracy by analysing specificity ranges in blood pressure and glucose levels will be useful to combat Diabetes Mellitus.

A data-driven optimization of large-scale dry port location using the hybrid approach of data mining and complex network theory

The paper proposes a two-stage approach that combines data mining and complex network theory to optimize the locations and service areas of dry ports in a large-scale inland transportation system. In the first stage, candidate locations of dry ports are weighted based on their eigenvector centrality in the complex network of association rules mined from a large amount of international transaction data. In the second phrase, dry port locations and their service areas are optimized using the gravity-based community structure. The method is validated in a real case study which optimizes a large-scale dry port network in Mainland China in the context of the Belt and Road Initiatives (BRI). As a result, optimal dry port locations include key transportation hubs that closely reflect the real BRI development plan, hence, the proposed approach is validated.

Болезнь Альцгеймера и искусственный интеллект: долговременная персонифицированная реабилитация и медико-социальное сопровождение

Modern technologies and tools for the rehabilitation of patients with Alzheimer’s disease have many potential applications for the treatment of dementia from diagnosis and evaluation to medical care, medical, social and economic support: from healthy aging, to accelerated and pathological aging of Homo sapiens. Doctor and neuroscientist: a modern solution to problems of rehabilitation ‘cognitive brain’ of Homo sapiens using on the one hand, tools and technologies of artificial intelligence, and with another — a multidisciplinary collaboration with clinical neurophysiologist ‘universal’ specialist in the field of neurology, psychiatry, psychotherapy, psychoanalysis and geriatrics. Systems biology, Biophysics, physiology, neurophysiology and highlight the multidimensional and combinatorial profiles of genetic, biological, pathophysiological and clinical biomarkers that reflects the heterogeneity of neurodegeneration, by means of modern efficient analysis tools to register and create comprehensive maps of the brain and recording of dynamic models in different systems: from molecules to neurons to brain regions. Bioinformatics, neuroimaging and neurophysiology of systems are aimed at calculating neural network models of the relationship between structure and dynamic function in brain networks. Structural and functional markers of the brain establish a link between clinical phenotypes and molecular pathophysiological mechanisms. Phenotypic variability is now considered one of the biggest problems in gerontology and geriatrics. MRI imaging to detect subtle changes in brain tissue and structure, fMRI imaging to measure changes in brain activity, and EEG to measure electrical activity have given clinicians many new insights into what happens in the brain in healthy and pathological aging. The paradigm of system neurophysiology is aimed at studying the fundamental principles of functioning of integrated neural systems through the integration and analysis of neural information recorded in a multimodal way (for example, fMRI and EEG), through computational modeling and combining data mining methods. The ultimate goal of systemic neurophysiology is to find out how signals are represented in neocortical networks and what role many different neural components play. Modern artificial intelligence technologies are capable of many things, including predicting Alzheimer's disease with the help of combined and hybrid neuroimaging, sequencing of a new generation, etc., in order to start timely and effective rehabilitation brain Homo sapiens.

An Analytical Model for Evaluating Social Security Schemes-A Focus on “Ayushman Bharat” Universal Health Scheme in India

The government initiated social security schemes in countries such as India, target a large proportion of the population to provide various types of benefits that involve a number of stakeholders. Such schemes are executed by a large number of transactions between the Government agencies and the other stakeholders on a real time basis, thus resulting in large data sets. Current research advancements in the domain of social security schemes include analysis of sequential activities and debt occurrences for such transactions at the national level only. It has been a challenge in recent times to monitor and evaluate the performance of such gigantic schemes which also involves financial decision making at different levels. This paper proposes an innovative frame-work that combines data mining strategies with actuarial techniques to evaluate one of the popular schemes in India, namely AB-PMJAY (“Ayushman Bharat–Pradhan Mantri Jan Arogya Yojana”) launched by the Government in 2018 at family level. In the proposed framework, the scheme has been divided into a number of sub-processes for which various data mining techniques such as, clustering, classification, anomaly detection and actuarial techniques for pricing are proposed to evaluate the scheme effective at micro level.

Candidate Genes and MiRNAs Linked to the Inverse Relationship Between Cancer and Alzheimer's Disease: Insights From Data Mining and Enrichment Analysis.

The incidence of cancer and Alzheimer’s disease (AD) increases exponentially with age. A growing body of epidemiological evidence and molecular investigations inspired the hypothesis of an inverse relationship between these two pathologies. It has been proposed that the two diseases might utilize the same proteins and pathways that are, however, modulated differently and sometimes in opposite directions. Investigation of the common processes underlying these diseases may enhance the understanding of their pathogenesis and may also guide novel therapeutic strategies. Starting from a text-mining approach, our in silico study integrated the dispersed biological evidence by combining data mining, gene set enrichment, and protein-protein interaction (PPI) analyses while searching for common biological hallmarks linked to AD and cancer. We retrieved 138 genes (ALZCAN gene set), computed a significant number of enriched gene ontology clusters, and identified four PPI modules. The investigation confirmed the relevance of autophagy, ubiquitin proteasome system, and cell death as common biological hallmarks shared by cancer and AD. Then, from a closer investigation of the PPI modules and of the miRNAs enrichment data, several genes (SQSTM1, UCHL1, STUB1, BECN1, CDKN2A, TP53, EGFR, GSK3B, and HSPA9) and miRNAs (miR-146a-5p, MiR-34a-5p, miR-21-5p, miR-9-5p, and miR-16-5p) emerged as promising candidates. The integrative approach uncovered novel miRNA-gene networks (e.g., miR-146 and miR-34 regulating p62 and Beclin1 in autophagy) that might give new insights into the complex regulatory mechanisms of gene expression in AD and cancer.

CROWD4EMS: A CROWDSOURCING PLATFORM FOR GATHERING AND GEOLOCATING SOCIAL MEDIA CONTENT IN DISASTER RESPONSE

Abstract. Increase in access to mobile phone devices and social media networks has changed the way people report and respond to disasters. Community-driven initiatives such as Stand By Task Force (SBTF) or GISCorps have shown great potential by crowdsourcing the acquisition, analysis, and geolocation of social media data for disaster responders. These initiatives face two main challenges: (1) most of social media content such as photos and videos are not geolocated, thus preventing the information to be used by emergency responders, and (2) they lack tools to manage volunteers contributions and aggregate them in order to ensure high quality and reliable results. This paper illustrates the use of a crowdsourcing platform that combines automatic methods for gathering information from social media and crowdsourcing techniques, in order to manage and aggregate volunteers contributions. High precision geolocation is achieved by combining data mining techniques for estimating the location of photos and videos from social media, and crowdsourcing for the validation and/or improvement of the estimated location. The evaluation of the proposed approach is carried out using data related to the Amatrice Earthquake in 2016, coming from Flickr, Twitter and Youtube. A common data set is analyzed and geolocated by both the volunteers using the proposed platform and a group of experts. Data quality and data reliability is assessed by comparing volunteers versus experts results. Final results are shown in a web map service providing a global view of the information social media provided about the Amatrice Earthquake event.

Spatiotemporal characteristics of green travel: A classification study on a public bicycle system

Abstract Understanding the characteristics of users and stations provides the foundation for a more efficient public bicycle system. Based on the real-time data of the Nanjing public bicycle system, we presented the spatiotemporal characteristics of users and stations combining data mining and geographic visualization. First, we analyzed users' gender, age, weekly flow, and time-segment flow, and classified the users into different types. In addition, we studied the cycling chains of certain users in details to understand the differences. Second, we analyzed the station distribution, station flow, station time-segment flow, and the surrounding environment, and studied the specific stations of different types to reveal the diverse characteristics. Moreover, we also explored the relationship between the user types and the station types. The results showed that public bicycles were mainly used for commuting or transferring, and social and economic activities around stations greatly influenced the use of public bicycles. However, the usage of the public bicycle system was still at a low level. Furthermore, different types of users had different cycling purposes, and different types of stations showed different characteristics of renting flow and returning flow. At last, we proposed different incentives and management measures for different types of users and stations.

Data-Driven Discovery of Full-Visible-Spectrum Phosphor

The development of extra-broadband phosphors is essential for next-generation illumination with better color experience. In this work, we report the discovery of the first-known Eu2+-activated full-visible-spectrum phosphor, Sr2AlSi2O6N:Eu2+, identified by combining data mining of high-throughput density functional theory calculations and experimental characterization. Excited by UV-light-emitting diodes (LEDs), Sr2AlSi2O6N:Eu2+ shows a superbroad emission with a bandwidth of 230 nm, the broadest emission bandwidth ever reported, and has excellent thermal quenching resistance (88% intensity at 150 °C). A prototype white LED utilizing only this full-visible-spectrum phosphor exhibits superior color quality (Ra = 97, R9 = 91), outperforming commercial tricolor phosphor-converted LEDs. These findings not only show great promise of Sr2AlSi2O6N:Eu2+ as a single white emitter but also open up in silico design of full-visible-spectra phosphor in a single-phase material to address the reabsorption energy loss in ...

Combining classification and regression for improving shear wave velocity estimation from well logs data

A shear wave is a critical property that helps to constrain rock properties such as lithology, pore fluid, and pore pressure. Nevertheless, valid and reliable estimation of shear wave velocity in highly heterogeneous reservoirs is still considered a challenge to be faced. In this paper, we propose a strategy for improving shear wave velocity estimation based on the combination of clustering and classification plus regression using conventional logs. The approach includes two steps. In step one, data clustering algorithms are applied to build a classifier model for identifying the distinct electrofacies (EF) based on similarity of well log responses. In step two, a distinct and specialized regression model is selected from more sophisticated methods (empirical models, statistical regression, virtual intelligence, and rock physics modeling) for determining the final shear wave velocity profile. The effectiveness of this proposed strategy is validated over a dataset from offset wells and their respective measured shear wave velocity data. The final results revealed that the proposed strategy that combines data mining task of clustering and classification plus regression, led to more uniform and better predictive performance of shear wave velocity (R2 = 0.97) in comparison with the use of stand-alone rock physics model (R2 = 0.70), virtual intelligence (R2 = 0.82), and statistical regression (R2 = 0.91). The silent features of this proposed strategy is that field-specific regression models are built based on log derived EF and geological information.

A Review of the Application of Machine Learning and Data Mining Approaches in Continuum Materials Mechanics

Machine learning tools represent key enablers for empowering material scientists and engineers to accelerate the development of novel materials, processes and techniques. One of the aims of using such approaches in the field of materials science is to achieve high-throughput identification and quantification of essential features along the process-structure-property-performance chain. In this contribution, machine learning and statistical learning approaches are reviewed in terms of their successful application to specific problems in the field of continuum materials mechanics. They are categorized with respect to their type of task designated to be either descriptive, predictive or prescriptive; thus to ultimately achieve identification, prediction or even optimization of essential characteristics. The respective choice of the most appropriate machine learning approach highly depends on the specific use-case, type of material, kind of data involved, spatial and temporal scales, formats, and desired knowledge gain as well as affordable computational costs. Different examples are reviewed involving case-by-case dependent application of different types of artificial neural networks and other data-driven approaches such as support vector machines, decision trees and random forests as well as Bayesian learning, and model order reduction procedures such as principal component analysis, among others. These techniques are applied to accelerate the identification of material parameters or salient features for materials characterization, to support rapid design and optimization of novel materials or manufacturing methods, to improve and correct complex measurement devices, or to better understand and predict fatigue behavior, among other examples. Besides experimentally obtained datasets, numerous studies draw required information from simulation-based data mining. Altogether, it is shown that experiment- and simulation-based data mining in combination with machine leaning tools provide exceptional opportunities to enable highly reliant identification of fundamental interrelations within materials for characterization and optimization in a scale-bridging manner. Potentials of further utilizing applied machine learning in materials science and empowering significant acceleration of knowledge output are pointed out.

Using combined multi-criteria decision-making and data mining methods for work zone safety: A case analysis

Work zone accidents are important concerns for transportation decision-makers. Therefore, knowledge of driving behaviors and traffic patterns are essential for identifying significant risk factors (RF) in work zones. Such knowledge can be difficult obtain in a field study without introducing new risks or driving hazards. This research uses integrated data mining and multi-criteria decision-making (MCDM) methods as part of a simulator-based case study of work zone logistics along a highway in Missouri. The research design incorporates k-mean clustering to cluster driving behavior trends, analytic network process (ANP) to determine weights for criteria that are most likely to impact work zones, and the Vlsekriterijumska Optimizacija I Kompromisno Resenje (VIKOR) method to rank the alternatives (clusters). Transportation engineers and decision makers can use results from this case study to identify driving populations most likely to engage in risky driving behaviors within work zones, and to provide guidance on effective work zone management.

A Data Mining Based Dispatching Rules Selection System for the Job Shop Scheduling Problem

Identifying the best Dispatching Rule in order to minimize makespan in a Job Shop Scheduling Problem is a complex task, since no Dispatching Rule is better than all others in different scenarios, making the selection of a most effective rule which is time-consuming and costly. In this paper, a novel approach combining Data Mining, Simulation, and Dispatching Rules is proposed. The aim is to assign in real-time a set of Dispatching Rules to the machines on the shop floor while minimizing makespan. Experiments show that the suggested approach is effective and reduces the makespan within a range of 1–44%. Furthermore, this approach also reduces the required computation time by using Data Mining to determine and assign the best Dispatching Rules to machines.

A Case Study of Operation Optimization on A Renewable Energy Building by E-CPS Method: From Both Sides of Supply and Demand

The energy system’s operation strategy has a significant influence on the system efficiency. The determination of the optimal operation strategy requires the effective application of a substantial amount of data accumulated from the operation or even forecasting process. However, data processing, which is not obtained based on the physical mechanism, may fail to explain, under certain boundary conditions. Therefore, using E-CPS to combine data mining and physical model to build a more efficient and accurate model or even a real-time model is vital and has received more attention in practical applications. This paper proposes a general process for constructing E-CPS and applies it to the simulation of building energy systems. The E-CPS results in a 70% reduction in the renewable energy waste rate of the building’s energy system and a 57% reduction in the purchase of electricity. Finally, it is concluded that the use of intelligent algorithm-based data mining combined with physical models to build E-CPS is practical and effective.