检索结果-内蒙古大学图书馆

International Conference on Mechatronics Engineering and Computing Technology (ICMECT)

作者： Chen Fuxing Xie Xusheng Jiangxi Normal Univ Coll Comp Informat Engn Nanchang Peoples R China

ISBN: (纸本)9783038351153

The query cost usually as an important criterion for a distributed database. The genetic algorithm is an adaptive probabilistic search algorithm, but the crossover and mutation probability usually keep a probability in traditional genetic algorithm. If the crossover probability keep a large value, the possibility of damage for genetic algorithm model is greater;In turn, if the crossover probability keep a small value, the search process will transform a slow processing or even stagnating. If the mutation probability keep a small value, a new individual can be reproduced difficultly;In turn, if the mutation probability keep a large value, the genetic algorithm will as a Pure random search algorithm. To solve this problem, proposed a improved genetic algorithm that multiple possibility of crossover and mutation based on k-means clustering algorithm. The experiment results indicate that the algorithm is effective.

关键词： distributed database genetic algorithm distributed query optimization k-means clustering algorithm

来源：评论

学校读者我要写书评

暂无评论

Detecting Anomaly Using the Scene Modeling Based on Time Delayed Statistical Data

Detecting Anomaly Using the Scene Modeling Based on Time Del...

引用

IEEE International Conference on Information and Automation

作者： Lixin Chen Huiwen Guo Min Wang Yen-Lun Chen Xinyu Wu Wei Feng Key Laboratory of Human-Machine Intelligence-Synergic Systems Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences

ISBN: (纸本)9781467391054

We propose a novel approach for the crowd anomaly detection in multiple cameras with non-overlapping view. In this paper, we refer to the activities of crowd in far-field scenes. Firstly, we present a model for learning all of the motion patterns under single camera view, which are regarded as the normal situation. In the surveillance region, we mark the entrances and exits under the single camera view and acquire the crowd flow model by the k-means clustering algorithm. Secondly, we analyze the crowd flow model based on the time delayed statistical data between two camera views. And then we acquire the relative location among the entrances and exits in the different regions. Thirdly, we analyze the crowd transferring probabilistic model on the global scene based on the log-likelihood function and Dirichlet distribution to detect the crowd anomaly. We set up the empirical threshold value of probability P_c. If the probability of detected model is less than P_c, the detected model is marked as the crowd anomaly. Our approach is evaluated on the simulated data set and the real data set in far-field scenes. Experimental results show the anomaly detection is precise.

关键词： Crowd anomaly detection Non-overlapping view k-means clustering algorithm Log-likelihood function Dirichlet distribution

来源：评论

学校读者我要写书评

暂无评论

An unsupervised feature extractionl method for classification of hyperspectral images 22

An unsupervised feature extractionl method for classificatio...

引用

22nd Iranian Conference on Electrical Engineering, ICEE 2014

作者： Imani, Maryam Ghassemian, Hassan Faculty of Electrical and Computer Engineering Tarbiat Modares University Iran

ISBN: (纸本)9781479944095

Hyperspectral images provide huge volume of spectral information for classification of land cover classes. Feature reduction plays an important role as a pre-processing step in classification of high dimensional data. Because of limited available training samples, unsupervised feature extraction is a proper selection for reduction of feature space. We propose an unsupervised feature extraction method in this paper that is called boundary clustering based feature extraction (BCFE). In the BCFE, at first using a clustering algorithm, data is clustered. We use the k-means clustering algorithm in this paper. After clustering, by training of SVM with using the obtained clusters, boundary samples of clusters are calculated. These boundary samples are used for discriminant analysis in the proposed feature extraction method. The experimental results on two real hyperspectral images show the advantage of BCFE in comparison with the most conventional feature extraction methods such as principal component analysis (PCA) and linear discriminant analysis (LDA). © 2014 IEEE.

关键词： feature extraction hyperspectral imaging image classification support vector machines unsupervised learning BCFE method k-means clustering algorithm LDA PCA SVM training boundary clustering based feature extraction high dimensional data classification hyperspectral image classification land cover class classification linear discriminant analysis principal component analysis spectral information unsupervised feature extraction method Accuracy Educational institutions Feature extraction Hyperspectral imaging Principal component analysis Support vector machines Training extraction-clustering-boundary feature sample-classification-hyperspectral image

来源：评论

学校读者我要写书评

暂无评论

Non-invasive method to analyse the risk of developing diabetic foot

引用

HEALTHCARE TECHNOLOGY LETTERS 2014年第4期1卷 109-113页

作者： Silva, Rebeca N. Ferreira, Ana C. B. H. Ferreira, Danton D. Barbosa, Bruno H. G. Univ Fed Lavras Dept Engn BR-37200000 Lavras MG Brazil Univ Fed Juiz de Fora Sch Nursing BR-36036900 Juiz De Fora MG Brazil

Foot complications (diabetic foot) are among the most serious and costly complications of diabetes mellitus. Amputation of all or part of a lower extremity is usually preceded by a foot ulcer. To prevent diabetic foot, an automatic non-invasive method to identify patients with diabetes who have a high risk of developing diabetic foot is proposed. To design the proposed method, information concerning social scope and self-care of 153 diabetic patients was presented to the k-means clustering algorithm, which divided the data into two groups: high risk and low risk of developing diabetic foot. In the operational stage, the Euclidian distance from the information vector to the centroids of each group of risk is used as criterion for classification. Both real and simulated data were used to evaluate the method in which promising results were achieved with accuracy of 0.97 + 0.06 for simulated data and 0.68 + 0.16 considering the classification of specialists as the gold standard for real data. The method requires a simple computational processing and can be useful for basic health units to triage diabetic patients helping the health-care team to reduce the number of cases of diabetic foot.

关键词： Euclidian distance k-means clustering algorithm automatic noninvasive method computational processing data classification diabetes mellitus diabetic foot diabetic patients diseases foot complications foot ulcer health care information vector medical diagnostic computing operational stage patient diagnosis pattern classification pattern clustering risk analysis self-care simulated data social scope

来源：评论

学校读者我要写书评

暂无评论

Comparison of clustering algorithms on generalized propensity score in observational studies: a simulation study

引用

JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION 2013年第12期83卷 2206-2218页

作者： Tu, Chunhao Jiao, Shuo koh, Woon Yuen Univ New England Coll Pharm Portland ME 04103 USA Fred Hutchinson Canc Res Ctr Div Publ Hlth Seattle WA 98109 USA Univ New England Dept Math Sci Biddeford ME 04005 USA

In observational studies, unbalanced observed covariates between treatment groups often cause biased inferences on the estimation of treatment effects. Recently, generalized propensity score (GPS) has been proposed to overcome this problem;however, a practical technique to apply the GPS is lacking. This study demonstrates how clustering algorithms can be used to group similar subjects based on transformed GPS. We compare four popular clustering algorithms: k-means clustering (kMC), model-based clustering, fuzzy c-means clustering and partitioning around medoids based on the following three criteria: average dissimilarity between subjects within clusters, average Dunn index and average silhouette width under four various covariate scenarios. Simulation studies show that the kMC algorithm has overall better performance compared with the other three clustering algorithms. Therefore, we recommend using the kMC algorithm to group similar subjects based on the transformed GPS.

关键词： fuzzy c-means clustering algorithm generalized propensity score k-means clustering algorithm model-based clustering algorithm observational studies partitioning around medoids algorithm

来源：评论

学校读者我要写书评

暂无评论

Frequency-hopping signals sorting based on underdetermined blind source separation

引用

IET COMMUNICATIONS 2013年第14期7卷 1456-1464页

作者： Sha, Zhi-Chao Huang, Zhi-Tao Zhou, Yi-Yu Wang, Feng-Hua Natl Univ Def Technol Coll Elect Sci & Engn Changsha 410073 Hunan Peoples R China

Most of the earlier approaches for frequency-hopping (FH) signals sorting with a single sensor do not adapt to synchronous network, whereas the multiple-sensor-based algorithms request the number of sensors must be more than that of signals. Since the number of sensors is limited in many practical applications, it is important to sort synchronous or asynchronous FH networks with as little as possible sensors. This study introduces the underdetermined blind source separation (UBSS) algorithm to solve this problem. Considering the time-frequency (TF) sparsity of FH signals, the problem is formulated as one of UBSS based on sparse TF representation. First, an improved k-means clustering algorithm is developed to estimate the mixing matrix, according to the TF properties of FH signals. Secondly, to separate more signals overlapped in the TF domain using given sensors, an improved subspace-based algorithm utilising the information of the comparative power is proposed. In the proposed method, the sparsity condition of the sources in the TF domain is relaxed, and the number of FH signals that exist at any TF point simultaneously is allowed to equal that of the sensors. Simulations demonstrate that the proposed method can separate FH signals efficiently and outperforms the previous methods.

关键词： blind source separation estimation theory frequency hop communication frequency hopping signal underdetermined blind source separation UBSS algorithm time-frequency sparsity sparse time-frequency representation k-means clustering algorithm mixing matrix estimation comparative power source sparsity condition time-frequency domain time-frequency point

来源：评论

学校读者我要写书评

暂无评论

A batch-based run-to-run process control scheme for semiconductor manufacturing

引用

IIE TRANSACTIONS 2013年第6期45卷 658-669页

作者： Wang, kaibo Han, kai Tsinghua Univ Dept Ind Engn Beijing 100084 Peoples R China

Run-to-Run (R2R) control has been widely used in semiconductor manufacturing to compensate for process disturbance and to improve quality. The traditional R2R controller only takes the process output in a previous run as its input and generates an optimal recipe for the next run. In a multistage semiconductor manufacturing process, variations in upstream stations are propagated to downstream stations. However, the information from upstream stations is not considered by existing controllers. In addition, most R2R processes have a limited capacity;the products must be processed in batches. Therefore, if the incoming materials could be grouped with small within-batch variations and large batch-to-batch variations and the recipes are customized for each batch to drive all batch averages toward the same target value, the output variation could be reduced and quality improved. A batch Exponentially Weighted Moving Average (EWMA) controller is proposed in this article. It employs a modified k-means algorithm to group the incoming materials into batches with a fixed and equal size while minimizing the within-batch variation. The controller then generates the control settings by taking both the batch information and the feedback quality information into account. Simulation studies show that the proposed controller could significantly reduce output variation and improve quality.

关键词： Batch-to-batch variation k-means clustering algorithm run-to-run process control semiconductor manufacturing within-batch variation

来源：评论

学校读者我要写书评

暂无评论

Dynamic Delegation Approach in the Electronic Evidence Storage of Cloud Platform

Dynamic Delegation Approach in the Electronic Evidence Stora...

引用

3rd International Conference on Applied Mechanics, Materials and Manufacturing (ICAMMM 2013)

作者： Wang, Yanfei Xie, Jun Xian, Zhengguang Capital Normal Univ Coll Informat Engn Beijing Peoples R China

ISBN: (纸本)9783037858882

The trustworthiness of the web resources is very important for secure cloud platform of electronic evidence storage calculation. In this paper, I analyzed the possibility of classes of delegation and trust level of authorization access. I used CGRS resource selection algorithm which is based on the k-means clustering algorithm in digital forensics analysis of cloud platform resources, and the algorithm is used to scalable environment for digital forensics analysis platform in substantive departments. The algorithm is proved effective and correct by the analysis and the experimental results.

关键词： Electronic evidence Cloud platform k-means clustering algorithm CGRS resource selection algorithm Dynamic Access Control

来源：评论

学校读者我要写书评

暂无评论

Collective Assessment of Arguments in an Online Intelligent Argumentation System for Collaborative Decision Support

Collective Assessment of Arguments in an Online Intelligent ...

引用

International Conference on Collaboration Technologies and Systems (CTS)

作者： Arvapally, Ravi Santosh Liu, Xiaoqing (Frank) Missouri Univ Sci & Technol Dept Comp Sci Rolla MO 65409 USA

ISBN: (纸本)9781467364041;9781467364034

We developed an intelligent argumentation system for collaborative decision support which allows stakeholders to exchange arguments, captures their rationale, and assesses aggregate thoughts on solution alternatives [1, 21]. In this article we propose and evaluate a method for computing aggregate thoughts of arguments on individual arguments, which represents individual thoughts, in an argumentation tree and further classifying arguments based on relationships between its aggregate thoughts and individual thoughts. A novel approach is developed to derive collective determination of an argument based on total support and attack that an argument receives in the argumentation tree. The collective determination value and the strength of an argument are used to analyze the relationship between aggregate thought and individual thought of arguments and cluster the arguments. The decision maker will be able to review clusters of arguments with opposing collective thoughts and supporting collective thoughts versus their individual thoughts. We have carried out an experiment to evaluate the proposed method, and the experimental results show that the proposed method is effective.

关键词： Collaborative decision support argumentation fuzzy logic aggregate thoughts individual thoughts k-means clustering algorithm intelligent argumenation system

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Satellite Image Segmentation using a Bivariate Beta type-II Mixture Model

Unsupervised Satellite Image Segmentation using a Bivariate ...

引用

IEEE 9th International Conference on Computational Cybernetics (ICCC)

作者： Ben Arab, Taher Zribi, Mourad Masmoudi, Afif Lab Informat Signal & Image Cote Opale LISIC EA 4 ULCO50 Rue Ferdinand Buisson BP 719 F-62228 Calais France Univ Sfax Fac Sci Lab Probabilities Statist Sfax Tunisia

ISBN: (纸本)9781479900633;9781479900602

This paper presents an unsupervised satellite color image segmentation approach based on Bivariate Beta type-II. Such a method could be considered as original since it uses a. k-means clustering algorithm in order to initialize the image class number. Moreover, it exploits a Bivariate Beta type-II for statistical distributions applying it for each class. Satellite image exploitation requires the use of different approaches, especially those based on the unsupervised statistical segmentation principle. Such approaches necessitate the definition of several parameters such as image class number, class variables estimation and mixture distributions. The use of statistical image attributes has allowed us to get convincing results, provided that we ensure under the condition of having an initialization step with appropriate statistical distributions. Bivariate Beta type-II associated with a. k-means clustering algorithm and Expectation-Maximization (EM) algorithm could be adapted to such a problem. For each image class, Bivariate Beta typeII attributes a specific distribution type according to different parameters. Different adapted algorithms (namely. k-means clustering algorithm, EM algorithm and Bivariate Beta type-II algorithm) are then applied to the satellite image segmentation problem. The efficiency of those combined algorithms is validated with the Mean Squared Errors (MSE), Signal to Noise Ratio (SNR) and Maximum Distance (MD).

关键词： Bivariate Beta type-II Expectation Maximization algorithm k-means clustering algorithm Mean Squared Errors Unsupervised color image segmentation

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：