Improved k-means clustering algorithm

In allusion to the disadvantage of having to obtain the number of clusters of data sets in advance and the sensitivity to selecting initial clustering centers in the k-means algorithm, an improved k-means clustering algorithm is proposed. First, the concept of a silhouette coefficient is introduced, and the optimal clustering number Kopt of a data set with unknown class information is confirmed by calculating the silhouette coefficient of objects in clusters under different K values. Then the distribution of the data set is obtained through hierarchical clustering and the initial clustering-centers are confirmed. Finally, the clustering is completed by the traditional k-means clustering. By the theoretical analysis, it is proved that the improved k-means clustering algorithm has proper computational complexity. The experimental results of IRIS testing data set show that the algorithm can distinguish different clusters reasonably and recognize the outliers efficiently, and the entropy generated by the algorithm is lower.

Study on the Application of K-Means Algorithm Implemented Hadoop Platform to the Library Work in Colleges and Universities

In this paper, the borrowing data of readers is analyzed and studied by taking K-Means algorithm as an example and implementing this algorithm in Hadoop calculation platform, and data mining technology is effectively and closely combined with personalized library service through the experimental data.

半监督学习算法在农用地分等中的应用

为了提高了土地评价模型的简易性、可解释性和准确性,以及克服传统土地评价模型中认为因素多的影响,提出利用关联规则挖掘算法从已知类别的训练样本提取其中的分类关联规则作为监督信息,结合非监督学习方法中的K-mean聚类算法,对大量未标定样本进行分类的半监督学习方法。该方法实现过程简单,分类准确率高,可推广性较强。对广东省土地资源的评价实验表明,利用半监督学习算法可得到较高的土地评价准确率94.0622%。

基于Chameleon算法和谱平分法的聚类新方法

在分析传统的聚类算法优越性和存在不足的基础上,基于Chameleon算法和谱平分法的思想提出了一种新的聚类方法。相比传统聚类算法而言此算法克服了如k-means算法、EM算法等传统聚类算法在聚类不为凸的样本空间时容易陷入局部最优的缺点,能在任意形状的样本空间上聚类,且收敛于全局最优解,并且可以降低噪声和离群点的影响,提高了算法的有效性。在UCI数据集和5个特殊的二维数据点组成的数据集上进行了实验,证明了本方法的有效性。

Research on control technology of elderly-assistant & walking-assistant robot based on tactile-slip sensation

In this paper,according to the old people's physical characteristics and their technical requirements for comfort and mastery when operating the robot,a control approach driven by tactile and slip senses is investigated to control the elderly-assistant & walking-assistant robot. First,on the basis of the proposed driving control system program of tactile and slip,a detection system of tactile and slip senses are designed. Based on the tactile and slip feature representation and extraction,an improved classification and recognition method is proposed which combines K-nearest neighbor (KNN) algorithm and K-means algorithm. And then,a robot control system based on TMS320F2812 is designed in this paper,including its hardware and software design. Then,a moving control method including the fuzzy adaptive control algorithm is presented for the walking-assistant robot to realize some different moving properties. At last,by the experimental verification in the walking-assistant robot,the research results show that the tactile and slip senses detection and recognition method is effective,and the whole control system has good feasibility and adaptability.

依托学科竞赛的计算机专业学生能力培养研究

在总结国内外知名大学生程序设计竞赛特点的基础上,提出依托学科竞赛的计算机专业学生能力培养模式:结合不同竞赛的特点,完善计算机专业课程讲授体系;构建师生协作和教学过程的质量保障体系。通过对近3年应届毕业生的就业状况和竞赛成绩进行统计和聚类分析,说明依托学科竞赛的学生能力培养模式是可行的。

P2P网络中的数据挖掘

在分析了现有分布式数据挖掘算法的运行机制和P2P技术具有无中心、不同步等特点的基础上,通过扩展经典K-mean算法的迭代过程,设计了一种能够用于P2P网络的分布式数据挖掘算法。该算法只需要在直接相连的节点间传递数据,并且能使每个节点上的数据按照全局聚类的结果聚合。最后用模拟实验验证了该算法的有效性。

Research on Clustering Analysis and Its Application in Customer Data Mining of Enterprise

The paper study improved K-means algorithm and establish indicators to classify customers according to RFM model. Experimental results show that, the new algorithm has good convergence and stability, it has better than single use of FKP algorithms for clustering. Finally the paper study the application of clustering in customer segmentation of mobile communication enterprise. It discusses the basic theory, customer segmentation methods and steps, the customer segmentation model based on consumption behavior psychology, and the segmentation model is successfully applied to the process of marketing decision support.

Applying memetic algorithm-based clustering to recommender system with high sparsity problem

A new recommendation method was presented based on memetic algorithm-based clustering. The proposed method was tested on four highly sparse real-world datasets. Its recommendation performance is evaluated and compared with that of the frequency-based, user-based, item-based, k-means clustering-based, and genetic algorithm-based methods in terms of precision, recall, and F1 score. The results show that the proposed method yields better performance under the new user cold-start problem when each of new active users selects only one or two items into the basket. The average F1 scores on all four datasets are improved by 225.0%, 61.6%, 54.6%, 49.3%, 28.8%, and 6.3% over the frequency-based, user-based, item-based, k-means clustering-based, and two genetic algorithm-based methods, respectively.

A New Method for Clustering Based on Development of Imperialist Competitive Algorithm

Clustering is one of the most widely used data mining techniques that can be used to create homogeneous clusters.K-means is one of the popular clustering algorithms that,despite its inherent simplicity,has also some major problems.One way to resolve these problems and improve the k-means algorithm is the use of evolutionary algorithms in clustering.In this study,the Imperialist Competitive Algorithm(ICA) is developed and then used in the clustering process.Clustering of IRIS,Wine and CMC datasets using developed ICA and comparing them with the results of clustering by the original ICA,GA and PSO algorithms,demonstrate the improvement of Imperialist competitive algorithm.

Proposal for Generation of the Three-Way Perceptual Map Using Non-metric Multidimensional Scaling with Clusters

The generation of a perceptual map via three-way multidimensional scaling allows analysts to see the separation of objects in Euclidean space. The MDSvarext method incorporates the objects＇ confidence regions in this analysis, allowing for statistical inference in the difference between objects, but the confidence regions that are generated are very large because of the inherent variability among the evaluators. One solution to this problem is cluster generation prior to the application of the MDSvarext method in order to obtain homogeneous subgroups and to achieve greater control of the variance. This work is relevant to studies of perception which usually evaluate the difference between objects or stimuli in the point of view of different people that judge this difference using several dimensions. This study investigated the possibility of using a K-means algorithm to generate subgroups before the MDSvarext method was applied, evaluating the process with two quality indicators, one Ex-Ante and one Ex-Post. The experiments were conducted based on simulation of judgment matrix of different objects in multiple dimensions being evaluated by several judges. In this experiment, the matrix used was a 10 objects, in 10 features, judged by 10 people. The results are promising as possible interpretations of the perceptual map and the indicators generated.

Intravascular Ultrasound Image Hard Plaque Recognition and Media-adventitia Border Detection

Intravascular ultrasound (IVUS) is a new technology for the diagnosis of coronary artery disease, and for the support of coronary intervention. IVUS image segmentation often encounters difficulties when plaque and acoustic shadow are present A novel approach for hard plaque recognition and media-adventitia border detection of IVUS images is presented in this paper. The IVUS images were first enhanced by a spatial-frequency domain filter that was constructed by the directional filter and histogram equalization. Then, the hard plaque was recognized based on the intensity variation within different regions that were obtained using the k-means algorithm. In the next step, a cost matrix representing the probability of the media-adventitia border was generated by combining image gradient, plaque location and image intensity. A heuristic graph-searching was applied to find the media-adventitia border from the cost matrix.Experiment results showed that the accuracy of hard plaque recognition and media-adventitia border detection was 89.94% and 95.57%, respectively. In conclusion,using hard plaques recognition could improve media-adventitia border detection in IVUS images.

A novel texture clustering method based on shift invariant DWT and locality preserving projection

We propose a novel texture clustering method. A classical type of(approximate) shift invariant discrete wavelet transform(DWT),dual tree DWT,is used to decompose texture images. Multiple signatures are generated from the obtained high-frequency bands. A locality preserving approach is applied subsequently to project data from high-dimensional space to low-dimensional space. Shift invariant DWT can represent image texture information efficiently in combination with a histogram signature,and the local geometrical structure of the dataset is preserved well during clustering. Experimental results show that the proposed method remarkably outperforms traditional ones.