一种基于用户偏好的协同过滤推荐算法

In this paper, a collaborative filtering recommendation algorithm based on user preference is proposed. First of all, the user similarity is calculated according to the length of the longest common subsequence of different user interest sequences and the num- ber of common subsequences, and then the similarity obtained by this algorithm is weighted and mixed with the similarity obtained by traditional collaborative filtering recommendation algorithm. Project recommendation is completed based on mixed similarity and the possible project score by target users is predicted. Finally, by comparing the average absolute error MAE values of three rec- ommendation algorithms in three data sets of Ciao, Flixster and MovieLens 100K, it is proved that the proposed user collaborative filtering recommendation algorithm (XQCF) has improved the accuracy of the recommendation system.     

Three‐stage multi‐innovation parameter estimation for an exponential autoregressive time‐series model with moving average noise by using the data filtering technique

This paper studies the data filtering‐based identification algorithms for an exponential autoregressive time‐series model with moving average noise. By means of the data filtering technique and the hierarchical identification principle, the identification model is transformed into three sub‐identification (Sub‐ID) models, and a filtering‐based three‐stage extended stochastic gradient algorithm is derived for identifying these Sub‐ID models. In order to improve the parameter estimation accuracy, a filtering‐based three‐stage multi‐innovation extended stochastic gradient (F‐3S‐MIESG) algorithm is developed by using the multi‐innovation identification theory. The simulation results indicate that the proposed F‐3S‐MIESG algorithm can work well.     

Training feedforward neural networks using multi-verse optimizer for binary classification problems

Discovering approximately recurrent motifs (ARMs) in timeseries is an active area of research in data mining. Exact motif discovery is defined as the problem of efficiently finding the most similar pairs of timeseries subsequences and can be used as a basis for discovering ARMs. The most efficient algorithm for solving this problem is the MK algorithm which was designed to find a single pair of timeseries subsequences with maximum similarity at a known length. This paper provides three extensions of the MK algorithm that allow it to find the top K similar subsequences at multiple lengths using both the Euclidean distance metric and scale invariant normalized version of it. The proposed algorithms are then applied to both synthetic data and real-world data with a focus on discovery of ARMs in human motion trajectories.     

多特征分层融合的相关滤波鲁棒跟踪

目的 为提高目标跟踪的鲁棒性,针对相关滤波跟踪中的多特征融合问题,提出了一种多特征分层融合的相关滤波鲁棒跟踪算法。方法 采用多通道相关滤波跟踪算法进行目标跟踪时,从目标和周围背景区域分别提取HOG（histogram of oriented gradient）、CN（color names）和颜色直方图3种特征。提出的分层融合算法首先采用自适应加权融合策略进行HOG和CN特征的特征响应图融合,通过计算特征响应图的平滑约束性和峰值旁瓣比两个指标得到融合权重。将该层融合结果与基于颜色直方图特征获得的特征响应图进行第2层融合时,采用固定系数融合策略进行特征响应图的融合。最后基于融合后的响应图估计目标的位置,并采用尺度估计算法估计得到目标更准确的包围盒。结果 采用OTB-2013（object tracking benchmark 2013）和VOT-2014（visual object tracking 2014）公开测试集验证所提跟踪算法的性能,在对多特征分层融合参数进行分析的基础上,与5种主流基于相关滤波的目标跟踪算法进行了对比分析。实验结果表明,本文算法的目标跟踪精度有所提高,其跟踪精度典型值比Staple算法提高了5.9%（0.840 vs 0.781）,同时由于有效地融合了3种特征,在多种场景下目标跟踪的鲁棒性优于其他算法。结论 提出的多特征分层融合跟踪算法在保证跟踪准确率的前提下,跟踪鲁棒性优于其他算法。当相关滤波跟踪算法采用了多个不同类型特征时,本文提出的分层融合策略具有一定的借鉴性。     

Prefix-projection global constraint and top-<Emphasis Type="Italic">k</Emphasis> approach for sequential pattern mining

Sequential pattern mining (SPM) is an important data mining problem with broad applications. SPM is a hard problem due to the huge number of intermediate subsequences to be considered. State of the art approaches for SPM (e.g., PrefixSpan Pei et al. 2001) are largely based on the pattern-growth approach, where for each frequent prefix subsequence, only its related suffix subsequences need to be considered, and the database is recursively projected into smaller ones. Many authors have promoted the use of constraints to focus on the most promising patterns according to the interests of the end user. The top-k SPM problem is also used to cope with the difficulty of thresholding and to control the number of solutions. State of the art methods developed for SPM and top-k SPM, though efficient, are locked into a rather rigid search strategy, and suffer from the lack of declarativity and flexibility. Indeed, adding new constraints usually amounts to changing the data-structures used in the core of the algorithm, and combining these new constraints often require new developments. Recent works (e.g. Kemmar et al. 2014; Négrevergne and Guns 2015) have investigated the use of Constraint Programming (CP) for SPM. However, despite their nice declarative aspects, all these modelings have scaling problems, due to the huge size of their constraint networks. To address this issue, we propose the Prefix-Projection global constraint, which encapsulates both the subsequence relation as well as the frequency constraint. Its filtering algorithm relies on the principle of projected databases which allows to keep in the variables domain, only values leading to a frequent pattern in the database. Prefix-Projection filtering algorithm enforces domain consistency on the variable succeeding the current frequent prefix in polynomial time. This global constraint also allows for a straightforward implementation of additional constraints such as size, item membership, regular expressions and any combination of them. Experimental results show that our approach clearly outperforms existing CP approaches and competes well with the state-of-the-art methods on large datasets for mining frequent sequential patterns, sequential patterns under various constraints, and top-k sequential patterns. Unlike existing CP methods, our approach achieves a better scalability.     

A fast projection algorithm for sequence data searching

In real life, data often appear in the form of sequences and this form of data is called sequence data. In this paper, a new definition on sequence similarity and a novel algorithm, Projection Algorithm, for sequence data searching are proposed. This algorithm is not required to access every datum in a sequence database. However, it guarantees that no qualified subsequence is falsely rejected. Moreover, the projection algorithm can be extended to match subsequences with different scales. With careful selection of parameters, most of the similar subsequences with different scales can be retrieved. We also show by experiments that the proposed algorithm can outperform the traditional sequential searching algorithm up to 96 times in terms of speed up.     

改进的曲率滤波强噪声图像去噪方法

目的 医学影像获取和视频监控过程中会出现一些恶劣环境,导致图像有许多强噪声斑点,质量较差。在处理强噪声图像时,传统的基于变分模型的算法,因需要计算高阶偏微分方程,计算复杂且收敛较慢;而隐式使用图像曲率信息的曲率滤波模型,在处理强噪声图像时,又存在去噪不完全的缺陷。为了克服这些缺陷,在保持图像边缘和细节特征的同时去除图像的强噪声,实现快速去噪,提出了一种改进的曲率滤波算法。方法 本文算法在隐式计算曲率时,通过半窗三角切平面和最小三角切平面的组合,用投影算子代替传统曲率滤波的最小三角切平面投影算子,并根据强噪声图像存在强噪声斑点的特征,修正正则能量函数,增添局部方差的正则能量,使得正则项的约束更加合理,提高了算法的去噪性能,从而达到增强去噪能力和保护图像边缘与细节的目的。结果 针对多种不同强度的混合噪声图像对本文算法性能进行测试,并与传统的基于变分法的去噪算法（ROF）和曲率滤波去噪等算法进行去噪效果对比,同时使用峰值信噪比（PSNR）和结构相似性（SSIM）作为滤波算法性能的客观评价指标。本文算法在对强噪声图像去噪处理时,能够有效地保持图像的边缘和细节特征,具备较好的PSNR和SSIM,在PSNR上比ROF模型和曲率滤波算法分别平均提高1.67 dB和2.93 dB,SSIM分别平均提高0.29和0.26。由于采用了隐式计算图像曲率,算法的处理速度与曲率滤波算法相近。结论 根据强噪声图像噪声特征对曲率滤波算法进行优化,改进投影算子和能量函数正则项,使得曲率滤波算法能够更好地适用于强噪声图像,实验结果表明,该方法与传统的变分法相比,对强噪声图像去噪效果显著。     

A novel image segmentation approach based on neutrosophic c-means clustering and indeterminacy filtering

This paper presents a novel image segmentation algorithm based on neutrosophic c-means clustering and indeterminacy filtering method. Firstly, the image is transformed into neutrosophic set domain. Then, a new filter, indeterminacy filter is designed according to the indeterminacy value on the neutrosophic image, and the neighborhood information is utilized to remove the indeterminacy in the spatial neighborhood. Neutrosophic c-means clustering is then used to cluster the pixels into different groups, which has advantages to describe the indeterminacy in the intensity. The indeterminacy filter is employed again to remove the indeterminacy in the intensity. Finally, the segmentation results are obtained according to the refined membership in the clustering after indeterminacy filtering operation. A variety of experiments are performed to evaluate the performance of the proposed method, and a newly published method neutrosophic similarity clustering (NSC) segmentation algorithm is utilized to compare with the proposed method quantitatively. The experimental results show that the proposed algorithm has better performances in quantitatively and qualitatively.

     

An adaptive GOP structure selection for haar-like MCTF encoding based on mutual information

In conventional motion compensated temporal filtering based wavelet coding scheme, where the group of picture structure and low-pass frame position are fixed, variations in motion activities of video sequences are not considered. In this paper, we propose an adaptive group of picture structure selection scheme, which the group of picture size and low-pass frame position are selected based on mutual information. Furthermore, the temporal decomposition process is determined adaptively according to the selected group of picture structure. A large amount of experimental work is carried out to compare the compression performance of proposed method with the conventional motion compensated temporal filtering encoding scheme and adaptive group of picture structure in standard scalable video coding model. The proposed low-pass frame selection can improve the compression quality by about 0.3–0.5 dB comparing to the conventional scheme in video sequences with high motion activities. In the scenes with un-even variation of motion activities, e.g. frequent shot cuts, the proposed adaptive group of picture size can achieve a better compression capability than conventional scheme. When comparing to adaptive group of picture in standard scalable video coding model, the proposed group of picture structure scheme can lead to about 0.2~0.8 dB improvements in sequences with high motion activities or shot cut.

Scheduling big data applications within advance reservation framework in optical grids

Providing QoS for big data applications requires a way to reserve computing and networking resources in advance. Within advance reservation framework, a multi-domain scheduling process is carried out in a top down hierarchical way across multiple hierarchical levels. This ensures that each domain executes intra-domain scheduling algorithm to co-schedule its own computing and networking resources while coordinating the scheduling at the inter-domain level. Within this process, we introduce two algorithms: iterative scheduling algorithm and K-shortest paths algorithm. We conducted a comprehensive performance evaluation study considering several metrics that reflect both grid system and grid user goals. The results demonstrated the advantages of the proposed scheduling process. Moreover, the results highlight the importance of using the iterative scheduling and K-shortest paths algorithms especially for data intensive applications.     

The use of explicit building blocks in evolutionary computation

This paper proposes a new algorithm to identify and compose building blocks. Building blocks are interpreted as common subsequences between good individuals. The proposed algorithm can extract building blocks from a population explicitly. Explicit building blocks are identified from shared alleles among multiple chromosomes. These building blocks are stored in an archive. They are recombined to generate offspring. The additively decomposable problems and hierarchical decomposable problems are used to validate the algorithm. The results are compared with the Bayesian optimisation algorithm, the hierarchical Bayesian optimisation algorithm, and the chi-square matrix. This proposed algorithm is simple, effective, and fast. The experimental results confirm that building block identification is an important process that guides the recombination procedure to improve the solutions. In addition, the method efficiently solves hard problems.     

方向小波域的选择性阈值SAR图像去噪

SAR图像去噪一直是SAR图像处理中一个具有特殊意义的研究课题。噪声抑制的关键是解决图像平滑与保持纹理之间的矛盾。提出了一种基于方向小波的选择性阈值SAR图像去噪算法。该算法利用方向小波的多方向框架对图像作12个方向的分解和变换。针对方向小波分解图像所产生的系数序列长度不同的特点,利用白噪声的置信区间,将不同长度的系数分成3组,对中间长度的系数序列采用统一阈值,对其他长度序列采用白噪声置信区间阈值处理。为了更好地保持图像细节信息,将每一尺度高频系数的方差中值作为噪声方差估计值。利用真实的SAR图像进行去噪试验,与几种经典的空域滤波和小波软阈值算法进行比较结果表明,该算法在平滑图像的同时更好地保持了图像本身的纹理信息,图像的视觉效果优于其他算法,等效视数和边缘保持指数分别提高了97和0.15。     

Effective database transformation and efficient support computation for mining sequential patterns

In this paper, we propose a novel algorithm for mining frequent sequences from transaction databases. The transactions of the same customers form a set of customer sequences. A sequence (an ordered list of itemsets) is frequent if the number of customer sequences containing it satisfies the user-specified threshold. The 1-sequence is a special type of sequences because it consists of only a single itemset instead of an ordered list, while the k-sequence is a sequence composed of k itemsets. Compared with the cost of mining frequent k-sequences (k ≥ 2), the cost of mining frequent 1-sequences is negligible. We adopt a two-phase architecture to find the two types of frequent sequences separately in order that the discovery of frequent k-sequences can be well designed and optimized. For efficient frequent k-sequence mining, every frequent 1-sequence is encoded as a unique symbol and the database is transformed into one constituted by the symbols. We find that it is unnecessary to encode all the frequent 1-seqences, and make full use of the discovered frequent 1-sequences to transform the database into one with a smaller size. For every k ≥ 2, the customer sequences in the transformed database are scanned to find all the frequent k-sequences. We devise the compact representation for a customer sequence and elaborate the method to enumerate all distinct subsequences from a customer sequence without redundant scans. The soundness of the proposed approach is verified and a number of experiments are performed. The results show that our approach outperforms the previous works in both scalability and execution time.     

传感器故障条件下的自适应UKF 算法

针对目标跟踪中传感器故障导致滤波发散或者滤波精度不高的问题, 提出一种自适应无迹卡尔曼滤波(UKF) 算法. 该算法在滤波过程中, 根据自适应估计原理引入自适应矩阵因子, 实时调整系统状态向量和量测新息向量的协方差, 以满足无迹卡尔曼滤波算法的最优性条件, 并采取措施对滤波发散的情况进行判断和抑制. 与传统UKF和已有自适应UKF算法相比, 该自适应UKF算法显著提高了滤波精度和数值稳定性, 且具有应对传感器故障的自适应能力. 仿真实验结果表明了所提出算法的有效性.

     

Inapproximability of maximal strip recovery

In comparative genomics, the first step of sequence analysis is usually to decompose two or more genomes into syntenic blocks that are segments of homologous chromosomes. For the reliable recovery of syntenic blocks, noise and ambiguities in the genomic maps need to be removed first. Maximal Strip Recovery (MSR) is an optimization problem proposed by Zheng, Zhu, and Sankoff for reliably recovering syntenic blocks from genomic maps in the midst of noise and ambiguities. Given d genomic maps as sequences of gene markers, the objective of MSR-d is to find d subsequences, one subsequence of each genomic map, such that the total length of syntenic blocks in these subsequences is maximized. For any constant d≥2, a polynomial-time 2d-approximation for MSR-d was previously known. In this paper, we show that for any d≥2, MSR-d is APX-hard, even for the most basic version of the problem in which all gene markers are distinct and appear in positive orientation in each genomic map. Moreover, we provide the first explicit lower bounds on approximating MSR-d for all d≥2. In particular, we show that MSR-d is NP-hard to approximate within Ω(d/logd). From the other direction, we show that the previous 2d-approximation for MSR-d can be optimized into a polynomial-time algorithm even if d is not a constant but is part of the input. We then extend our inapproximability results to several related problems including CMSR-d, δ-gap-MSR-d, and δ-gap-CMSR-d.     

基于VMD循环随机跳跃状态网络的时间序列长期预测

由于混沌系统的演化规律复杂,直接对混沌时间序列进行长期预测通常难以达到较好的效果.针对此问题,利用变分模态分解方法将混沌时间序列转化为一系列特征子序列,利用排列熵评估选取子序列个数的合理性,保证特征子序列包含了原序列长期演化趋势.此外,提出一种改进的确定性循环跳跃状态网络作为子序列的预测模型,该网络模型中的储备池采用单向环状连接和双向随机跳跃的拓扑结构,能够避免储备池确定连接结构造成的预测精度较低和随机连接造成网络的不稳定性问题.通过所提出模型对时间序列进行长期预测,采用多种评估手段对预测结果进行分析, 表明所提出模型对于长期预测具有较大的优势.     

BASE: A bacteria foraging algorithm for cell formation with sequence data

A cellular manufacturing system (CMS) is considered an efficient production strategy for batch type production. A CMS relies on the principle of grouping machines into machine cells and grouping parts into part families on the basis of pertinent similarity measures. The bacteria foraging algorithm (BFA) is a newly developed computation technique extracted from the social foraging behavior of Escherichia coli (E. coli) bacteria. Ever since Kevin M. Passino invented the BFA, one of the main challenges has been employment of the algorithm to problem areas other than those for which the algorithm was proposed. This research work studies the first applications of this emerging novel optimization algorithm to the cell formation (CF) problem considering the operation sequence. In addition, a newly developed BFA-based optimization algorithm for CF based on operation sequences is discussed. In this paper, an attempt is made to solve the CF problem, while taking into consideration the number of voids in the cells and the number of inter-cell travels based on operational sequences of the parts visited by the machines. The BFA is suggested to create machine cells and part families. The performance of the proposed algorithm is compared with that of a number of algorithms that are most commonly used and reported in the corresponding scientific literature, such as the CASE clustering algorithm for sequence data, the ACCORD bicriterion clustering algorithm and modified ART1, and using a defined performance measure known as group technology efficiency and bond efficiency. The results show better performance of the proposed algorithm.     

A blind audio watermarking algorithm by logarithmic quantization index modulation

In this paper we present a blind audio watermarking algorithm based on the vector norm and the logarithmic quantization index modulation (LQIM) in the wavelet domain, integrating the robustness of the vector norm with the imperceptibility of the logarithmic quantization index modulation based on μ-Law (or mu-Law) companding. Firstly μ-Law companding is adopted to transform the vector norm of the segmented wavelet approximation components of the original audio signal. And then a binary image scrambled by the chaotic sequence as watermark is embedded in the transformed domain with a uniform quantization scheme. Experimental results demonstrate that even if the capacity of the proposed algorithm is high, up to 102.4 bps, this algorithm can still maintain a high quality of the audio signal, and achieve a better performance, such as imperceptibility, robustness and complexity in comparison with the uniform quantization based algorithms against common attacks. What’s more, it can resist amplitude scaling attack effectively.     

Parametric calibration of speed-density relationships in mesoscopic traffic simulator with data mining

Speed-density relationships are used by mesoscopic traffic simulators to represent traffic dynamics. While classical speed-density relationships provide useful insights into the traffic dynamics problem, they may be restrictive for such applications. This paper addresses the problem of calibrating speed-density relationship parameters using data mining techniques, and proposes a novel hierarchical clustering algorithm based on K-means clustering. By combining K-means with agglomerative hierarchical clustering, the proposed new algorithm is able to reduce early-stage errors inherent in agglomerative hierarchical clustering resulted in improved clustering performance. Moreover, in order to improve the precision of parametric calibration, densities and flows are utilized as variables. The proposed approach is tested against sensor data captured from the 3rd Ring Road of Beijing. The testing results show that the performance of our algorithm is better than existing solutions.     

基于形状特征k-d树的多维时间序列相似搜索

多维时间序列是信息系统中一类重要的数据对象,相似搜索是其应用的一个核心.两个序列(子序列)相似度加以比较的常用方法是:将序列(子序列)转换成空间中的曲线,然后计算曲线间的欧几里德距离.这种方法的主要缺陷是它仅考虑了序列(子序列)间的整体距离关系,而不能体现它们自身的局部变化.针对此问题,提出了一种新的可应用于多维时间序列的快速相似搜索方法.该方法将序列(子序列)的局部变化特性与检索结构(k-d树)结合起来,使得在搜索k-d树的同时实现了序列(子序列)的局部变化匹配,从而极大地提高了查询效率和正确率.实验结果表明了算法的有效性.