基于键盘行为数据的用户身份识别

用户击键行为作为一种生物特征,具有采集成本低、安全性高的特点。然而,现有的研究方法和实验环境都是基于实验室数据,并不适用于极度不平衡的真实数据。比如,在实验室数据上效果出色的分类算法在真实数据上却无法应用。针对此问题,提出了基于真实击键行为数据的用户识别算法。该方法将聚类算法和距离算法结合起来,通过比较新来的击键行为和历史击键行为相似度以实现用户识别。实验结果表明,该算法在100名用户的3015条真实击键记录组成的数据集上准确率达到88.22%,在投入实际应用后,随着样本集的增大算法的准确率还可以进一步提升。     

云环境下基于数据流的k-means聚类算法

k-means算法是一种 最常用的基于划分的聚类算法。传统的集中式k-means算法已不能适应当前呈爆炸式增长的数据规模,设计分布式k-means算法成为了目前亟需解决的问题。现有分布式k-means算法基于MapReduce计算框架且没有考虑初始聚类中心的影响。由于每个MapReduce任务均需要读写分布式文件系统,导致MapReduce不能有效表达多个任务之间的依赖关系,因此提出了一种基于数据流的计算框架,该框架建立在MapReduce之上,将数据处理过程按照数据流图建模。在该框架的基础上,提出了一种高效的k-means算法,它采用基于多次采样的初始聚类中心选取方法来实现负载均衡及减少迭代次数。实验结果表明,该算法的可扩展性较好,且效率比现有算法高。     

基于主题短语的搜索引擎结果聚类

为了解决搜索引擎检索结果中的主题混杂现象,帮助用户快速准确地定位到有价值的信息,提出基于主题短语的搜索引擎结果聚类方法。首先从检索结果中提取查询词并与相邻词语组成主题短语,建立包含高频独立词语及主题短语的混合向量空间模型,同时引入同义词词林对特征项进行语义扩充,最后采用改进的k-means聚类算法对搜索结果进行聚类,并为各个类别提取类别标签。实验结果表明,该算法能有效提高聚类结果的准确率。     

GPU并行加速的均值偏移算法

为克服mean shift算法计算复杂度高、运行速度慢的缺点,提出一种基于GPU的快速mean shift算法.首先使用k-means算法对图像像素进行预分类,之后在预分类、下采样后缩小的数据集上进行mean shift聚类,以有效地降低算法复杂度.此外,借助GPU的通用计算功能对k-means和mean shift分别进行并行了处理.实验结果表明,通过对图像进行预处理,有效地提高了几何模板查找在强噪声、低信噪比图像中的识别率;同时,改进后的mean shift算法的运行速度提高了近40倍,满足了高速机器视觉检测的实时性要求.     

数据驱动的同调机组分群及主动解列技术

新型电力系统发生区域性故障时系统的安全稳定性问题日益突出,主动解列作为防止故障扩散的有效手段在系统稳定研究中逐渐获得关注。从同调机组分群技术和解列断面搜索技术两个方面对主动解列开展研究,结合时频分析、特征提取和图论等方法,提出基于改进聚类算法的同调机组分群技术和基于多层图分割的解列断面搜索技术,并在IEEE-39节点标准测试系统中验证了所提方法的适用性。     

Case-based expert system using wavelet packet transform and kernel-based feature manipulation for engine ignition system diagnosis

Whenever there is any fault in an automotive engine ignition system or changes of an engine condition, an automotive mechanic can conventionally perform an analysis on the ignition pattern of the engine to examine symptoms, based on specific domain knowledge (domain features of an ignition pattern). In this paper, case-based reasoning (CBR) approach is presented to help solve human diagnosis problem using not only the domain features but also the extracted features of signals captured using a computer-linked automotive scope meter. CBR expert system has the advantage that it provides user with multiple possible diagnoses, instead of a single most probable diagnosis provided by traditional network-based classifiers such as multi-layer perceptions (MLP) and support vector machines (SVM). In addition, CBR overcomes the problem of incremental and decremental knowledge update as required by both MLP and SVM. Although CBR is effective, its application for high dimensional domains is inefficient because every instance in a case library must be compared during reasoning. To overcome this inefficiency, a combination of preprocessing methods, such as wavelet packet transforms (WPT), kernel principal component analysis (KPCA) and kernel K-means (KKM) is proposed. Considering the ignition signals captured by a scope meter are very similar, WPT is used for feature extraction so that the ignition signals can be compared with the extracted features. However, there exist many redundant points in the extracted features, which may degrade the diagnosis performance. Therefore, KPCA is employed to perform a dimension reduction. In addition, the number of cases in a case library can be controlled through clustering; KKM is adopted for this purpose. In this paper, several diagnosis methods are also used for comparison including MLP, SVM and CBR. Experimental results showed that CBR using WPT and KKM generated the highest accuracy and fitted better the requirements of the expert system.     

Experience with a Hybrid Processor: K-Means Clustering

We discuss hardware/software co-processing on a hybrid processor for a compute- and data-intensive multispectral imaging algorithm, k-means clustering. The experiments are performed on two models of the Altera Excalibur board, the first using the soft IP core 32-bit NIOS 1.1 RISC processor, and the second with the hard IP core ARM processor. In our experiments, we compare performance of the sequential k-means algorithm with three different accelerated versions. We consider granularity and synchronization issues when mapping an algorithm to a hybrid processor. Our results show that speedup of 11.8X is achieved by migrating computation to the Excalibur ARM hardware/software as compared to software only on a Gigahertz Pentium III. Speedup on the Excalibur NIOS is limited by the communication cost of transferring data from external memory through the processor to the customized circuits. This limitation is overcome on the Excalibur ARM, in which dual-port memories, accessible to both the processor and configurable logic, have the biggest performance impact of all the techniques studied.     

Automatic localization of craniofacial landmarks for assisted cephalometry

In this paper we propose a system for localization of cephalometric landmarks. The process of localization is carried out in two steps: deriving a smaller expectation window for each landmark using a trained neuro-fuzzy system (NFS) then applying a template-matching algorithm to pin point the exact location of the landmark. Four points are located on each image using edge detection. The four points are used to extract more features such as distances, shifts and rotation angles of the skull. Limited numbers of representative groups that will be used for training are selected based on k-means clustering. The most effective features are selected based on a Fisher discriminant for each feature set. Using fuzzy linguistics if-then rules, membership degree is assigned to each of the selected features and fed to the FNS. The FNS is trained, utilizing gradient descent, to learn the relation between the sizes, rotations and translations of landmarks and their locations. The data for training is obtained manually from one image from each cluster. Images whose features are located closer to the center of their cluster are used for extracting data for the training set. The expected locations on target images can then be predicted using the trained FNS. For each landmark a parametric template space is constructed from a set of templates extracted from several images based on the clarity of the landmark in that image. The template is matched to the search windows to find the exact location of the landmark. Decomposition of landmark shapes is used to desensitize the algorithm to size differences. The system is trained to locate 20 landmarks on a database of 565 images. Preliminary results show a recognition rate of more than 90%.     

基于免疫原理的RBF网络Carrousel氧化沟系统预报

为提高Carrousel氧化沟系统水质模拟的精度,利用免疫算法结合RBF神经网络进行建模,并加入k-means聚类对样本进行预处理。以某污水处理中心两年生产数据进行实验,测得出水TN的预报误差0.195 6,出水TP的预报误差0.145 6,较仅用RBF网络均有很大提高,证明该方法可以应用于Carrousel氧化沟系统的在线实时预测。     

基于多分类SVM的石榴叶片病害检测方法

石榴是陕西临潼广泛种植的水果作物之一。石榴的生产力由于其果实、茎和叶中各种类型的疾病引起的感染而降低,叶片病害主要由细菌、真菌、病毒等引起。疾病是限制水果生产的一个主要因素,疾病往往难以控制,如果没有准确的疾病诊断,就不能在适当的时间采取适当的控制行动。图像处理技术是植物叶片病害检测和分类中广泛应用的技术之一,旨在利用支持向量机分类技术对石榴叶片病害进行检测和分类。首先用K均值聚类法分割出病变区域,然后提取颜色和纹理特征,最后采用LSVM（线性支持向量机）分类技术对叶片病害类型进行检测。所提出的系统可以成功地检测和分类所检查的疾病,准确率为89.55%。