Cluster Analysis In Data Mining Research Articles

Wind Turbine Fire Prevention System Using Fuzzy Rules and WEKA Data Mining Cluster Analysis

With the rapid expansion of the supply of renewable energy in accordance with the global energy transition policy, the wind power generation industry is attracting attention. Subsequently, various wind turbine control technologies have been widely developed and applied. However, there is a lack of research on optimal pitch control, which detects wind direction and changes the rotation angle of the blade in real time. In areas where the wind speed is not strong, such as South Korea, it is necessary to maintain the optimal angle in real time so that the rotating surface of the blade can face the wind direction. In this study, optimal pitch control was performed through real-time analysis of wind speed, direction, and temperature, which is the core of wind turbine maintenance, using fuzzy rules using FIS (Fuzzy Interface System) and WEKA data mining cluster analysis techniques. In order to prevent fires caused by the over-current of wind turbines, over-current control methods such as VCB (Vacuum Circuit Breaker) utilization, prototype utilization such as a modular MCB (Main Circuit Breaker) incorporating VI (Vacuum Interrupter), and vacuum degree change analysis methods using a PD (Partial Discharge) signal were proposed. The optimal control technique for wind turbine parts and facilities was put forth after judging and predicting the annual average wind distribution suitable for wind power generation using HRWPRM (Korea’s High-Resolution Wind Power Resource Maps). Finally, the various wind turbine control methods carried out in this study were confirmed through computer simulation, such as remote diagnosis and early warning issuance, prediction of power generation increase and decrease situation, and automatic analysis of wind turbine efficiency.

EXPLORING TOURIST EXPENDITURES OF THE GJIROKASTRA DISTRICT IN ALBANIA: A CLUSTER ANALYSIS

Tourism expenditures are an important indicator for measuring the impact of tourism on a destination. Information taken from the micro data of tourism expenditures, together with other socio-demographic characteristics of visitors, provides important data on the tourist demand for a destination. Visitor costs in combination with other variables can be used to profile the demand for tourist goods and services in the destination area. In this paper, the daily expenses of visitors in the Gjirokastra district of Albania are analysed. The Gjirokastra district is one of the more preferred regions and attracts a considerable number of domestic and foreign visitors, because it offers a diversity tourist attractions and services. For the purposes of this study, a survey of 200 visitors was conducted. The survey results were analysed with descriptive statistics via a cluster analysis. The survey provides detailed data on the daily expenses of visitors, divided into six main categories of a typical travel budget. Through the use of data mining cluster analysis, we identified and analysed the tourists’ daily expenses based on the purpose of their visit to the area of Gjirokaster.

College students’ Network behavior Using data mining and feature analysis

Teachers may use advanced analytics to rapidly and correctly understand undergraduate behavior trends, especially when it comes to identifying undergraduate groupings that need to be focused on at a later time. This study uses data mining cluster analysis to analyze the constituent behavior of 3,245 undergraduates in a specific level ‘B’ institution’s college network. According to the data, there are four different undergraduate groups with different Web access features, with 350 participants using the accomplishments and other variables of their success have an influence on these students. As a result of this research, we were able to collect data on undergraduate college network activity, which may be used to aid in the development of academic advising management.

Research on Customer Relationship of Power Marketing Based on Clustering Model of Internet

After entering the new century, information technology and Internet technology have developed rapidly, and the application of network technology has become more and more widely used. In the power marketing relationship, the adoption of new technologies has become increasingly important for the management of customer relationship. This paper combines the concept of difference degree in data mining clustering analysis algorithm to design an evaluation tool suitable for this model. The degree of difference in cluster analysis of data mining is used to describe the similarity between two objects. When the attribute values of the description object are different, the calculation method of the difference degree is also different. In the field of clustering analysis problems, a method for calculating the degree of difference has been proposed for the problem of two-dimensional variable high-dimensional sparse clustering.

Clustering Algorithm and Its Application in Data Mining

Clustering analysis is one of the main research directions in data mining. At present, it has gone deep into all fields and made good progress. Aiming at the role of clustering analysis in data mining, a clustering analysis algorithm and its application in data mining are proposed. Through literature comparative analysis method, the basic concepts of cluster analysis are expounded in detail, and the classical algorithms in cluster analysis are discussed. The basic realization process of clustering K-means algorithm is analyzed and an example simulation is carried out. The research shows that this algorithm has strong universality and can be applied to most data analysis sites, providing a theoretical basis for timely detection and analysis of large amounts of data.

Cross-efficiency evaluation in the presence of flexible measures with an application to healthcare systems.

In recent years, most countries around the world have struggled with the consequences of budget cuts in health expenditure, obliging them to utilize their resources efficiently. In this context, performance evaluation facilitates the decision-making process in improving the efficiency of the healthcare system. However, the performance evaluation of many sectors, including the healthcare systems, is, on the one hand, a challenging issue and on the other hand a useful tool for decision- making with the aim of optimizing the use of resources. This study proposes a new methodology comprising two well-known analytical approaches: (i) data envelopment analysis (DEA) to measure the efficiencies and (ii) data science to complement the DEA model in providing insightful recommendations for strategic decision making on productivity enhancement. The suggested method is a first attempt to combine two DEA extensions: flexible measure and cross-efficiency. We develop a pair of benevolent and aggressive scenarios aiming at evaluating cross-efficiency in the presence of flexible measures. Next, we perform data mining cluster analysis to create groups of homogeneous countries. Organizing the data in similar groups facilitates identifying a set of benchmarks that perform similarly in terms of operating conditions. Comparing the benchmark set with poorly performing countries we can obtain attainable goals for performance enhancement which will assist policymakers to strategically act upon it. A case study of healthcare systems in 120 countries is taken as an example to illustrate the potential application of our new method.

Man Power Allotment in Industry Sector Using Kmean Algorithm Special Reference On Rane Trw Steering Systems

Staff allotment is a crucial job in industry sector. Human resource plays leading rule to achieve success of the industry. The Data mining is analytical method to allot the right job for right man. This is not easiest job to match the work schedule in large industrial sector. There is a large databases have to be maintain each and every department. There must be a need for perfect knowledge discovery model to retrieve the information about the staff. From that the management can make perfect schedule. The data mining approaches will make suitable to make the work schedule. how can be Data mining techniques and k mean algorithm process the effective job will be the theme of this article.
 Data mining may be regarded as an evolving approach to data analysis in very large databases that could become a useful tool to management professionals. Data mining involves extracting knowledge based on patterns of data in very large databases. Yet, data mining goes beyond simply performing data analysis on large data sets. Organizations that employ thousands of employees and track a multitude of employment-related information might find valuable information patterns contained within their databases to provide insights in such areas as employee retention and compensation planning. To develop the staff planning and allotment, the k mean clustering algorithm can be used for this job. K mean is a method is popularly for cluster analysis in data mining .k means clustering aims to part ion n observations into k clusters in which each observation belongs to the cluster with nearest mean serving as prototype of cluster. k mean algorithm is can be grouped employees as a different cluster with nearest mean.

Advanced Approach of Material Region Detections on Fibre-Reinforced Concrete CT-Scans

Detections of material regions on CT-scans of solids are commonly treated manually by an expert. Although such manual detections have many advantages, some amount of human error is also incorporated. Moreover, expert opinions may vary significantly. We present an application of the k-means++ clustering as an alternative option to manual way of material area detections. k-means++ clustering is derived from k-means (the method of vector quantization, originally from signal processing), popular for cluster analysis in data mining and image processing communities. The algorithm s main advantages are its simple implementation and fast convergence to a local optimum of an objective function. We benchmark the suggested approach on transverse CT-scans of a fibre-reinforced concrete solid. Moreover, we introduce a technique for processing air distribution, such that the appropriate pixels detected as the pixels of air are converted into pixels representing concrete. The technique is based on the connected component algorithm. Benchmark and results of proposed method conclude the paper.

Spatial equilibrium of housing provident fund in China based on data mining cluster analysis

Housing provident fund is an important guarantee for the livelihood of people and also an important policy for the improvement of national housing condition. Its development relates to system imputation, extraction, loan business, GDP, workers' disposable income and other factors. This paper chooses the housing provident fund in 31 provinces in China as the research project, constructing the compound evaluation index system by unsupervised learning clustering algorithm to obtain the national average housing provident fund development distribution. Further in-depth analysis of the differences among provinces provides scientific basis for the national development strategy of housing provident fund.

Cluster Analysis of Electrical Behavior

In this paper, we apply clustering analysis of data mining into power system. We adapt K-means clustering algorithm to analyze customer load, analyzing similar behavior between customer of electricity, and we adapt principal component analysis to get the clustering result visible, Simulation and analysis using matlab, and this well verify cluster rationality. The conclusion of this paper can provide important basis to the peak for the power system, stable operation the power system security.

Factors associated with low fitness in adolescents--a mixed methods study.

BackgroundFitness and physical activity are important for cardiovascular and mental health but activity and fitness levels are declining especially in adolescents and among girls. This study examines clustering of factors associated with low fitness in adolescents in order to best target public health interventions for young people.Methods1147 children were assessed for fitness, had blood samples, anthropometric measures and all data were linked with routine electronic data to examine educational achievement, deprivation and health service usage. Factors associated with fitness were examined using logistic regression, conditional trees and data mining cluster analysis. Focus groups were conducted with children in a deprived school to examine barriers and facilitators to activity for children in a deprived community.ResultsUnfit adolescents are more likely to be deprived, female, have obesity in the family and not achieve in education. There were 3 main clusters for risk of future heart disease/diabetes (high cholesterol/insulin); children at low risk (not obese, fit, achieving in education), children ‘visibly at risk’ (overweight, unfit, many hospital/GP visits) and ‘invisibly at risk’ (unfit but not overweight, failing in academic achievement). Qualitative findings show barriers to physical activity include cost, poor access to activity, lack of core physical literacy skills and limited family support.ConclusionsLow fitness in the non-obese child can reveal a hidden group who have high risk factors for heart disease and diabetes but may not be identified as they are normal weight. In deprived communities low fitness is associated with non-achievement in education but in non-deprived communities low fitness is associated with female gender. Interventions need to target deprived families and schools in deprived areas with community wide campaigns.

Community detection by graph Voronoi diagrams

Accurate and efficient community detection in networks is a key challenge for complex network theory and its applications. The problem is analogous to cluster analysis in data mining, a field rich in metric space-based methods. Common to these methods is a geometric, distance-based definition of clusters or communities. Here we propose a new geometric approach to graph community detection based on graph Voronoi diagrams. Our method serves as proof of principle that the definition of appropriate distance metrics on graphs can bring a rich set of metric space-based clustering methods to network science. We employ a simple edge metric that reflects the intra- or inter-community character of edges, and a graph density-based rule to identify seed nodes of Voronoi cells. Our algorithm outperforms most network community detection methods applicable to large networks on benchmark as well as real-world networks. In addition to offering a computationally efficient alternative for community detection, our method opens new avenues for adapting a wide range of data mining algorithms to complex networks from the class of centroid- and density-based clustering methods.

Application of Silence Customer Segmentation in Securities Industry Based on Fuzzy Cluster Algorithm

Due to the unique advantage of soft classification, fuzzy cluster analysis can realize the customer segmentation and improve customer relationship according to the uncertainty and fuzziness of customer’s behavior. Based on the research approach, this study applies the fuzzy cluster algorithm to silence customers’ segmentation in securities industry and to identify the customers with similar characteristics and value. This study proposes a multi-dimensional segmentation model for silence customers with 8 indicators of industry characteristics based on the theory of Customer Relationship Management (CRM), and carries out an empirical research on real customer data and transaction data by the fuzzy clustering analysis of data mining, then identifies different groups of silence customers with similar characteristics. Finally, some marketing strategies are put forward in correspondence with different traits and preferences with the purpose of waking them up.

Fuzzy Clustering Algorithm for Silent Customer Segmentation in Securities Industry

Due to the unique advantage of soft classification, fuzzy cluster analysis can realize the customer segmentation and improve customer relationship according to the uncertainty and fuzziness of customer’s behavior. Based on the research approach, this study applies the fuzzy cluster algorithm to silent customers’ segmentation in securities industry and to identify the customers with similar characteristics and value. This study proposes a multi-dimensional segmentation model for silent customers with 8 indicators of industry characteristics based on the theory of customer relationship management (CRM), and carries out an empirical research on real customer data and transaction data by the fuzzy clustering analysis of data mining, then identifies different groups of silent customers with similar characteristics. Finally, some marketing strategies are put forward in correspondence with different traits and preferences with the purpose of waking them up.

The Smart Urban Planning Based on the Cluster Analysis Algorithm

The so-called 3S technology, including remote sensing technology, geographic information system, global positioning system, can be used to collect vast amounts of data for the city. The urban planners can access to knowledge from the cluster data by using cluster analysis in data mining. This paper presents an improved DBSCAN clustering analysis algorithm, E-DBSCAN, which is then applied to the urban planning in the New Town of Yuetang in Xiangtan city and provides aid for the planners.

Clustering Analysis Based on Data Mining Applications

Abstract. In this paper, a clustering algorithm based on data mining technology applications, the use of the extraction mode noise characteristics amount and pattern recognition algorithms, extraction and selection of the characteristic quantities of the three types of mode, carried out under the same working conditions data mining clustering analysis ultimately satisfying recognition.

Data mining cluster analysis on the influence of health factors in Casemix data

Background This study explores potential data mining applications in the Casemix context, which is expected to yield effective and efficient health care services. The objective of work focuses on determining hidden relevant patterns which can’t be processed by human capabilities all alone. California Drug and Alcohol treatment Assessment (CALDATA) of administrative type database can be relevant study for the medical diagnosis in usage of alcohol and drugs for patients admitted and discharged during the stay in hospital to discover knowledge for recovery process.

Research in Social Network Based on Rough Set Clustering Algorithm

This paper uses an improved algorithm based on rough set and K-means clustering to do research in social network, and it solves the problem of vague border. Social networks’ community is similar to the cluster analysis in data mining. However ,when the K-means clustering algorithm is used in social network,there are several disadvantages as following: how to determine the value of K,and the relations among community node and the community. This paper solves the disadvantages using rough set clustering ideas. This method is mainly used to find overlapping communities, and it can multi-anglely reflect the social network information better. The algorithm is applied to experiment data, and results show that this method improves the accuracy of community division obviously compared with other algorithms.

EIDBSCAN: An Extended Improving DBSCAN algorithm with sampling techniques

Cluster analysis in data mining and knowledge discovery is an essential business application. This investigation describes a new clustering approach named EIDBSCAN that extends expansion seed selection into a sampling-based DBSCAN clustering algorithm. Additionally, the proposed algorithm may reduce eight Marked Boundary Objects to add expansion seeds according to far centrifugal force, which increases coverage. Our experimental results reveal that the proposed EIDBSCAN yields more accurate clustering results. In addition, in all the cases we studied, the proposed approach has a lower execution time cost than several existing well-known approaches, such as DBSCAN, IDBSCAN and KIDBSCAN clustering algorithms.

A honeybee-mating approach for cluster analysis

Cluster analysis, which is the subject of active research in several fields, such as statistics, pattern recognition, machine learning, and data mining, is to partition a given set of data or objects into clusters. K-means is used as a popular clustering method due to its simplicity and high speed in clustering large datasets. However, K-means has two shortcomings. First, dependency on the initial state and convergence to local optima. The second is that global solutions of large problems cannot be found with reasonable amount of computation effort. In order to overcome local optima problem lots of studies done in clustering. Over the last decade, modeling the behavior of social insects, such as ants and bees, for the purpose of search and problem solving has been the context of the emerging area of swarm intelligence. Honeybees are among the most closely studied social insects. Honeybee mating may also be considered as a typical swarm-based approach to optimization, in which the search algorithm is inspired by the process of marriage in real honeybee. Neural networks algorithms are useful for clustering analysis in data mining. This study proposes a two-stage method, which first uses self-organizing feature maps (SOM) neural network to determine the number of clusters and then uses honeybee mating optimization algorithm based on K-means algorithm to find the final solution. We compared proposed algorithm with other heuristic algorithms in clustering, such as GA, SA, TS, and ACO, by implementing them on several well-known datasets. Our finding shows that the proposed algorithm works better than others. In order to further demonstration of the proposed approach’s capability, a real-world problem of an Internet bookstore market segmentation based on customer loyalty is employed.