一种高效的多变量时间序列相似查询算法

为了高效地执行多变量时间序列（MTS）相似查询,提出一种基于距离的索引结构（Dbis）相似查询算法。采用主成分分析方法对MTS数据进行降维处理;聚类MTS主成分序列,选择每类质心作为参考点;依据参考点将每类变换到一维空间,这样可以利用B+ 树结构进行索引查询;MTS序列比较相似采用的是扩展的Frobenius范数（Eros）。通过对股票数据集实验验证了Dbis算法的高效性。     

面向噪声数据的时间序列相似性搜索研究

为了减少噪声数据对查询最优序列的影响,避免Euclidean距离对形态的敏感性,以及要求序列等长的缺点,提出了面向噪声数据的时间序列相似性搜索算法.运用SPC方法去除序列中的噪声数据;采用DTW距离作为度量函数,使用规范化方法使序列处于相同的分辨率下;采用LB_ Keogh下界函数对候选序列集合进行筛选.仿真实验结果表明,该算法在阈值较小时,对含有噪声数据序列的匹配能力较强.     

一种基于分形和相似性查找的非平稳时间序列符号化表示法

传统的时间序列表示方法均在不同程度上采用了对数据的约简手段,从而破坏了时间序列的非线性和分形这些重要的本质特征,也就使得时间序列的相似性匹配误差加大。提出一种高精度的随机非平稳时间序列表示方法FSPA,该方法将分形理论和R/S方法应用到现有的时间序列表示方法中,既保留了时间序列的非线性和分形的重要特征,同时也实现了维度的约简。实验分别在合成数据和实际数据上进行,结果表明,该方法具有更高的精度且需要较少的存储空间。     

基于斜率偏离的时间序列相似性搜索方法研究*

摘要：针对数据挖掘领域中时间序列的相似性度量问题,提出了基于斜率偏离度量的浮动索引相似性搜索算法。在斜率偏离度量的基础上建立分箱标志,通过浮动索引方法有效实现高维多元时间序列的准确索引,用实际的飞行数据进行仿真验证获得好的效果,证明这种方法的有效性和可靠性。     

一种高效的基于相似性查找时间序列的位符号化表示方法*

到目前为止能够计算字符化时间序列的距离度量的方法很少,为此,提出了一种新的字符化的时间序列表示方法BSAP。该方法既能进行维度约简又允许在符号化后的时间序列表示法上定义距离度量。实验分别在合成数据和实际数据上进行,实验表明该方法具有更高的运算效率且需要较少的空间。     

Optimal algorithms for the online time series search problem

In the problem of online time series search introduced by El-Yaniv et al. (2001) [1], a player observes prices one by one over time and shall select exactly one of the prices on its arrival without the knowledge of future prices, aiming to maximize the selected price. In this paper, we extend the problem by introducing profit function. Considering two cases where the search duration is either known or unknown beforehand, we propose two optimal deterministic algorithms respectively. The models and results in this paper generalize those of El-Yaniv et al. (2001) [1].     

在时间序列相似性问题中滑动窗口的确定

作为一个非平凡命题,大多数时间序列相似性查找方法都涉及到了对原数据的维度简约.在保持原序列中有效信息量的同时,尽量降低计算复杂度是这些算法的关键.讨论滑动窗口在时间序列相似性降维技术中的实际应用,从中发现确定自适应滑动窗口大小的一种新方法.通过对时序特征值分布函数的挖掘,发现时间序列中的若干有效点,从而确定一组合适的滑动窗口大小,并根据序列变化的来决定最佳的滑动窗口.     

基于涨落模式的时间序列相似性度量研究

时间序列形态相似性挖掘是目前时间序列数据挖掘研究的热点,然而由于时间序列数据背后真实系统的复杂性,加上观测条件的影响,时间序列会呈现多种相似性变形,如振幅伸缩、振幅漂移、线性漂移等。相似性变形并不会改变序列的形态特征,但现有的ED、DTW和L_p距离等相似性度量算法均不能有效支持识别各类相似性变形。本文首次提出涨落模式(FP)的概念,以涨落模式保存原序列的趋势变化信息,利用最长公共子序列算法计算涨落模式的相似度,消除振幅伸缩、振幅漂移和线性漂移等对相似性挖掘带来的影响,实现基于涨落模式的时间序列相似性度量。设置仿真数据集检验FP相似性度量的相似性变形支持性,同时在真实数据集上进行分类,依据分类准确性对算法鲁棒性进行评估,验证了本文提出的基于涨落模式的相似性度量算法在各类相似性形变上的有效支持性。     

基于三角不等式的时间序列相似性搜索算法

由于传统的时序相似性度量方式不满足距离三角不等式关系,影响后续的相似性搜索及关联规则的获取,在时序符号化的基础上,提出一种满足三角不等式的符号化距离度量方式。与MINDIST_PAA_SAX和Sym_PAA_SAX度量方式进行比较,其结果表明,该度量方式在异常检测和相似性查询上具有较好的优越性。实验结果表明,该方法在相似性搜索及关联规则的获取方面具有更高的可信度。     

Adaptive tree similarity learning for image retrieval

Learning-enhanced relevance feedback is one of the most promising and active research directions in content-based image retrieval in recent years. However, the existing approaches either require prior knowledge of the data or converge slowly and are thus not coneffective. Motivated by the successful history of optimal adaptive filters, we present a new approach to interactive image retrieval based on an adaptive tree similarity model to solve these difficulties. The proposed tree model is a hierarchical nonlinear Boolean representation of a user query concept. Each path of the tree is a clustering pattern of the feedback samples, which is small enough and local in the feature space that it can be approximated by a linear model nicely. Because of the linearity, the parameters of the similartiy model are better learned by the optimal adaptive filter, which does not require any prior knowledge of the data and supports incremental learning with a fast convergence rate. The proposed approach is simple to implement and achieves better performance than most approaches. To illustrate the performance of the proposed approach, extensive experiments have been carried out on a large heterogeneous image collection with 17,000 images, which render promising results on a wide variety of queries.An early version of part of the system was reported in Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition 2001.     

基于分布式架构的时间序列局部相似检测算法

基于动态时间规整算法思想的CrossMatch算法可以用来解决序列间的部分相似问题,但是由于算法时间空间复杂度过高,需要消耗大量的计算资源,因此无法应用于长序列之间的计算。针对以上问题,提出了一个基于分布式平台上的时间序列局部相似性检测算法。将CrossMatch算法实现在了分布式框架上,解决了计算资源不足的问题。首先需要对序列进行切分,分别放置在不同的节点上;其次,各节点分别处理各自序列的相似部分;最后,通过对结果进行汇总并拼接,找出序列间的局部相似。实验结果表明,该算法在准确性上和CrossMatch相近,在时间上也有提升。改进后的分布式算法不仅解决了单机无法处理的长序列计算问题,而且可以通过增加并行计算节点数提高运行速度。     

基于特征点分段的多元时间序列相似性搜索

现有的各种多元时间序列相似性搜索方法难以准确高效地完成搜索任务。提出了一种基于特征点分段的多元时间序列相似性搜索算法,提取所定义的用于分段的特征点,分段后将原时间序列转化为模式序列,该模式序列能够很好地保留原序列的全局形状特征,再用分层匹配的方法进行相似性搜索。实验结果表明,该方法能够有效刻画序列的全局形状特征,通过分层匹配保留局部的相似性,同时提高搜索准确率。     

利用反馈的时序模式挖掘算法研究

针对时序数据相似性挖掘方法进行研究，提出一种利用反馈的时序数据相似性挖掘算法，由用户赋予各初始范围查询得到的相似序列相应的权值，通过反馈与给定序列叠加产生新的查询序列，再次进行范围查询，获得相似序列，将该算法用于某钢铁企业的电力负荷时序数据，计算结果表明了算法的有效性。     

An adaptive technique for content-based image retrieval

We discuss an adaptive approach towards Content-Based Image Retrieval. It is based on the Ostensive Model of developing information needs—a special kind of relevance feedback model that learns from implicit user feedback and adds a temporal notion to relevance. The ostensive approach supports content-assisted browsing through visualising the interaction by adding user-selected images to a browsing path, which ends with a set of system recommendations. The suggestions are based on an adaptive query learning scheme, in which the query is learnt from previously selected images. Our approach is an adaptation of the original Ostensive Model based on textual features only, to include content-based features to characterise images. In the proposed scheme textual and colour features are combined using the Dempster-Shafer theory of evidence combination. Results from a user-centred, work-task oriented evaluation show that the ostensive interface is preferred over a traditional interface with manual query facilities. This is due to its ability to adapt to the user's need, its intuitiveness and the fluid way in which it operates. Studying and comparing the nature of the underlying information need, it emerges that our approach elicits changes in the user's need based on the interaction, and is successful in adapting the retrieval to match the changes. In addition, a preliminary study of the retrieval performance of the ostensive relevance feedback scheme shows that it can outperform a standard relevance feedback strategy in terms of image recall in category search.     

基于分层动态时间弯曲的序列相似性度量方法研究

为了更好地体现时间序列的形态特征,并探索更适合于较长时间序列之间相似性度量的方法,在动态时间弯曲算法的基础上进行改进,提出了基于分层动态时间弯曲的序列相似性度量方法。对时间序列进行多层次分段,并从分段中均匀抽取相对应的层次分段子序列,然后将层次分段子序列抽象为三维空间的点（反映了分段子序列的均值、长度和趋势）进行相似性度量,最后综合各个层次的相似性度量作为结果。实验表明,在参数设置合理的情况下,此方法能获得较高的序列相似性度量准确度和效率。     

基于PPR的煤矿瓦斯监测数据相似搜索方法

研究基于时间序列相似搜索技术的煤矿瓦斯涌出分析新途径,提出基于PPR的煤矿瓦斯监测数据相似搜索方法。实验采用玉华煤矿的真实煤矿瓦斯监测数据,评价指标为信息损失量及相似查询效率。与基于离散傅立叶变换(DFT)和离散小波变换(DWT)的时间序列相似搜索算法的对比实验显示：在相同压缩比下,3种方法的信息损失相近;但是基于PPR的相似搜索算法的平均查询效率分别比基于DFT和基于DWT方法高32%和34%。因此PPR算法适合用于瓦斯监测数据相似搜索。     

Accelerated queries of time series databases

It has been previously proposed that a query to a database of time signals can be accelerated by searching over partial data in Fourier space. It is proposed here that such queries can be accelerated further by employing composite Fourier filtering. In particular, many fractional power filters are trained on sets of vectors from the database. In single comparisons information about the entire set of vectors is available. Query times are shortened mainly due to the ability to examine several vectors in the database simultaneously.     

Relevance feedback for category search in music retrieval based on semantic concept learning

Traditional content-based music retrieval systems retrieve a specific music object which is similar to what a user has requested. However, the need exists for the development of category search for the retrieval of a specific category of music objects which share a common semantic concept. The concept of category search in content-based music retrieval is subjective and dynamic. Therefore, this paper investigates a relevance feedback mechanism for category search of polyphonic symbolic music based on semantic concept learning. For the consideration of both global and local properties of music objects, a segment-based music object modeling approach is presented. Furthermore, in order to discover the user semantic concept in terms of discriminative features of discriminative segments, a concept learning mechanism based on data mining techniques is proposed to find the discriminative characteristics between relevant and irrelevant objects. Moreover, three strategies, the Most-Positive, the Most-Informative, and the Hybrid, to return music objects concerning user relevance judgments are investigated. Finally, comparative experiments are conducted to evaluate the effectiveness of the proposed relevance feedback mechanism. Experimental results show that, for a database of 215 polyphonic music objects, 60% average precision can be achieved through the use of the proposed relevance feedback mechanism.

基于提前终止的加速时间序列弯曲算法

动态时间弯曲(DTW)距离是时间序列相似搜索的一种重要距离度量,但其精确计算是一个性能瓶颈。针对此问题,提出一种名为EA_DTW的方法用于加速DTW距离的精确计算,该方法在计算累积距离矩阵中每个方格的距离时都判断其是否超过阈值,一旦超过则提前终止其余相关方格的距离计算;并对EA_DTW的过程进行了理论分析。实验对比表明,EA_DTW能够提高DTW的计算效率,在阈值与DTW距离相比较小时更加明显。     

Evaluate time delay from sensor's data by trend similarity search

Time delay estimation is a general issue in both signal processing and process control fields. Neither offline step impulse response-based methods nor least squares methods in control field estimate time delay directly from the real running data. Although the methods for signal processing directly evaluate the delay from signals, such as correlation calculation, coherence analysis and least mean square methods, they are mainly suitable for two signals only different at a time delay part and an attenuation factor. In this article, an estimation method is proposed which is directly based on the real running input and output data of a control plant. The input and output signals of a plant show raw monotony from each other in many cases. According to this feature, we estimate the delay by comparing the trend of two signals. Furthermore, it is extended to an adaptive method for estimating piecewise time-varying delay by sliding window and forgetting factor. The experiments on real plant show the good performances of our methods. The simulation experiments demonstrate that our basic method performs better than CCF or coherence analysis for the nonlinear plant and the adaptive one performs better than least mean square methods for the signals with transfer function except time delay.