图形技术分析的统计刻画

本文在Lo等(2000)研究的基础上,给出了基于局部极值点刻画的4种技术形态的数学定义,并基于局部多项式估计方法给出了这些形态的自动识别算法。应用所提出的算法对我国沪市A股进行了图形技术分析的实证研究。研究结果表明本文定义的技术形态包含一般统计意义上显著的附加信息,从而证明了图形技术分析的有效性。     

An efficient rough feature selection algorithm with a multi-granulation view

Feature selection is a challenging problem in many areas such as pattern recognition, machine learning and data mining. Rough set theory, as a valid soft computing tool to analyze various types of data, has been widely applied to select helpful features (also called attribute reduction). In rough set theory, many feature selection algorithms have been developed in the literatures, however, they are very time-consuming when data sets are in a large scale. To overcome this limitation, we propose in this paper an efficient rough feature selection algorithm for large-scale data sets, which is stimulated from multi-granulation. A sub-table of a data set can be considered as a small granularity. Given a large-scale data set, the algorithm first selects different small granularities and then estimate on each small granularity the reduct of the original data set. Fusing all of the estimates on small granularities together, the algorithm can get an approximate reduct. Because of that the total time spent on computing reducts for sub-tables is much less than that for the original large-scale one, the algorithm yields in a much less amount of time a feature subset (the approximate reduct). According to several decision performance measures, experimental results show that the proposed algorithm is feasible and efficient for large-scale data sets.     

Partial surface matching by using directed footprints

In this paper we present a new technique for partial surface and volume matching of images in three dimensions. In this problem, we are given two objects in 3-space, each represented as a set of points, scattered uniformly along its boundary or inside its volume. The goal is to find a rigid motion of one object which makes a sufficiently large portion of its boundary lying sufficiently close to a corresponding portion of the boundary of the second object. This is an important problem in pattern recognition and in computer vision, with many industrial, medical, and chemical applications. Our algorithm is based on assigning a directed footprint to every point of the two sets, and locating all the pairs of points (one of each set) whose undirected components of the footprints are sufficiently similar. The algorithm then computes for each such pair of points all the rigid transformations that map the first point to the second, while making the respective direction components of their footprints coincide. A voting scheme is employed for computing transformations which map significantly large number of points of the first set to points of the second set. Experimental results on various examples are presented and show the accurate and robust performance of our algorithm.     

Minimum classification error training in example based speech and pattern recognition using sparse weight matrices

The Minimum Classification Error (MCE) criterion is a well-known criterion in pattern classification systems. The aim of MCE training is to minimize the resulting classification error when trying to classify a new data set. Usually, these classification systems use some form of statistical model to describe the data. These systems usually do not work very well when this underlying model is incorrect. Speech recognition systems traditionally use Hidden Markov Models (HMM) with Gaussian (or Gaussian mixture) probability density functions as their basic model. It is well known that these models make some assumptions that are not correct. In example based approaches, these statistical models are absent and are replaced by the pure data. The absence of statistical models has created the need for parameters to model the data space accurately. For this work, we use the MCE criterion to create a system that is able to work together with this example based approach. Moreover, we extend the locally scaled distance measure with sparse, block diagonal weight matrices resulting in a better model for the data space and avoiding the computational load caused by using full matrices. We illustrate the approach with some example experiments on databases from pattern recognition and with speech recognition.     

Pattern search ranking and selection algorithms for mixed variable simulation-based optimization

The class of generalized pattern search (GPS) algorithms for mixed variable optimization is extended to problems with stochastic objective functions. Because random noise in the objective function makes it more difficult to compare trial points and ascertain which points are truly better than others, replications are needed to generate sufficient statistical power to draw conclusions. Rather than comparing pairs of points, the approach taken here augments pattern search with a ranking and selection (R&S) procedure, which allows for comparing many function values simultaneously. Asymptotic convergence for the algorithm is established, numerical issues are discussed, and performance of the algorithm is studied on a set of test problems.     

子句集的神经网络归结

给出基于神经网络的归结方法。首先将子句集S表示为δ形式，并且用算子对(⊙， )引入两种类型的神经元；然后用这两种神经元构造子句集S的神经网络结构；而后给出基于子句集的神经网络的归结算法；最后证明了该算法的完备性，并用实例进行了验证。     

An approach to a consistent qualification of decision procedures in probabilistic information processing

Pattern recognition seems to be a rather unique field of interwoven logical inference and decision theory applications. The existence of hundreds of theoretical and real pattern recognition devices forms an ideal basis for research on the structures of various approaches and their comparison. The task of pattern recognition is to select a hypotheses out of a set (e.g.: figures 0, …, 9) on the basis of given data (e.g. the black and white points of a digitized picture). There exists an ideal classifier to solve this problem as the theorem of Bayes provides a logically perfect connection between the input data and the result. But as the so called Bayes-machine proves completely unpractical for real purposesit is “approximated” by more or less complex “real” decision procedures.Thus the theorem of Bayes provides a starting point for the application of statistical considerations and information theory to the analysis of the structures of real decision procedures. The results allow a rather consistent and simple comparison of most decision procedures and provide a tool to estimate the performance of a given procedure in a given environment. The results apply not only to pattern recognition but also to many other fields such as imminence analysis and medical diagnosis.     

An analysis of convergence for a learning version of the subspace method

The learning subspace method of pattern recognition has been earlier introduced by Kohonen et al. in a speech recognition application, where the phonemes to be classified are given as spectral representations. In that method, the class subspaces are updated recursively using special rotation matrices, which depend on the training vectors entering one at a time. Here the learning algorithm based on these operators is represented in a general mathematical form, and almost sure convergence is shown to a given criterion that is a function of the statistics of the training set as well as of a set of nonrandom but free parameters. The proof employs current techniques in stochastic approximation theory. For illustration, the resulting classification criterion is then applied to a concrete pattern recognition situation with suitably chosen parameter values.     

基于SVM理论的一种新的数据分类方法

基于 SVM分类器在模式识别问题中有独特的优势 ,本文通过对标准 SVM模型的改造 ,提出了一种新的简单的数据分类方法 .理论分析和实验表明 ,该方法与标准 SVM分类方法相比具有处理大规模数据识别的能力且保持较高的样本识别率 ,节省存储空间等优势 .     

Design of Trend-Identification Algorithms with Learning (Algebraic Approach)

Section 1 presents the main questions associated with the application of the algebraic approach to pattern recognition and classification. The answers to these questions constitute a problem-oriented theory of this application domain. In Sections 2 and 3, the listed questions are used to construct a solvability theory for the design of trend-identification algorithms. We formally describe the initial information set (algorithm inputs), the final information set (algorithm outputs), the precedents, and the additional constraints. Solvability and regularity criteria are derived for the problem of construction of trend-identification algorithms.     

Approximated algorithms for the minimum dilation triangulation problem

The complexity status of the minimum dilation triangulation (MDT) problem for a general point set is unknown. Therefore, we focus on the development of approximated algorithms to find high quality triangulations of minimum dilation. For an initial approach, we design a greedy strategy able to obtain approximate solutions to the optimal ones in a simple way. We also propose an operator to generate the neighborhood which is used in different algorithms: Local Search, Iterated Local Search, and Simulated Annealing. Besides, we present an algorithm called Random Local Search where good and bad solutions are accepted using the previous mentioned operator. For the experimental study we have created a set of problem instances since no reference to benchmarks for these problems were found in the literature. We use the sequential parameter optimization toolbox for tuning the parameters of the SA algorithm. We compare our results with those obtained by the OV-MDT algorithm that uses the obstacle value to sort the edges in the constructive process. This is the only available algorithm found in the literature. Through the experimental evaluation and statistical analysis, we assess the performance of the proposed algorithms using this operator.     

Decision making with imprecise parameters

We analyze the impact of imprecise parameters on performance of an uncertainty-modeling tool presented in this paper. In particular, we present a reliable and efficient uncertainty-modeling tool, which enables dynamic capturing of interval-valued clusters representations sets and functions using well-known pattern recognition and machine learning algorithms. We mainly deal with imprecise learning parameters in identifying uncertainty intervals of membership value distributions and imprecise functions. In the experiments, we use the proposed system as a decision support tool for a production line process. Simulation results indicate that in comparison to benchmark methods such as well-known type-1 and type-2 system modeling tools, and statistical machine-learning algorithms, proposed interval-valued imprecise system modeling tool is more robust with less error.     

监督模糊模式识别交叉迭代模型

从模糊模式识别概念出发，建立一种以决策者经验、偏好为监督，在方案优属等级识别过程中确定最佳目标权重和方案优属度的监督模糊模式识别交叉迭代算法，该算法集成了决策偏好信息完全未知、部分未知、完全已知的主客观权重识别方法。并严格证明了该算法的局部收敛性。     

Studying the effect of using low-discrepancy sequences to initialize population-based optimization algorithms

In this paper, we investigate the use of low-discrepancy sequences to generate an initial population for population-based optimization algorithms. Previous studies have found that low-discrepancy sequences generally improve the performance of a population-based optimization algorithm. However, these studies generally have some major drawbacks like using a small set of biased problems and ignoring the use of non-parametric statistical tests. To address these shortcomings, we have used 19 functions (5 of them quasi-real-world problems), two popular low-discrepancy sequences and two well-known population-based optimization methods. According to our results, there is no evidence that using low-discrepancy sequences improves the performance of population-based search methods.     

Local generalized quadratic distance metrics: application to the <Emphasis Type="Italic">k</Emphasis>-nearest neighbors classifier

Finding the set of nearest neighbors for a query point of interest appears in a variety of algorithms for machine learning and pattern recognition. Examples include k nearest neighbor classification, information retrieval, case-based reasoning, manifold learning, and nonlinear dimensionality reduction. In this work, we propose a new approach for determining a distance metric from the data for finding such neighboring points. For a query point of interest, our approach learns a generalized quadratic distance (GQD) metric based on the statistical properties in a “small” neighborhood for the point of interest. The locally learned GQD metric captures information such as the density, curvature, and the intrinsic dimensionality for the points falling in this particular neighborhood. Unfortunately, learning the GQD parameters under such a local learning mechanism is a challenging problem with a high computational overhead. To address these challenges, we estimate the GQD parameters using the minimum volume covering ellipsoid (MVCE) for a set of points. The advantage of the MVCE is two-fold. First, the MVCE together with the local learning approach approximate the functionality of a well known robust estimator for covariance matrices. Second, computing the MVCE is a convex optimization problem which, in addition to having a unique global solution, can be efficiently solved using a first order optimization algorithm. We validate our metric learning approach on a large variety of datasets and show that the proposed metric has promising results when compared with five algorithms from the literature for supervised metric learning.     

A genetic algorithm for the unrelated parallel machine scheduling problem with sequence dependent setup times

In this work a genetic algorithm is presented for the unrelated parallel machine scheduling problem in which machine and job sequence dependent setup times are considered. The proposed genetic algorithm includes a fast local search and a local search enhanced crossover operator. Two versions of the algorithm are obtained after extensive calibrations using the Design of Experiments (DOE) approach. We review, evaluate and compare the proposed algorithm against the best methods known from the literature. We also develop a benchmark of small and large instances to carry out the computational experiments. After an exhaustive computational and statistical analysis we can conclude that the proposed method shows an excellent performance overcoming the rest of the evaluated methods in a comprehensive benchmark set of instances.     

A correlative classifiers approach based on particle filter and sample set for tracking occluded target

Target tracking is one of the most important issues in computer vision and has been applied in many fields of science, engineering and industry. Because of the occlusion during tracking, typical approaches with single classifier learn much of occluding background information which results in the decrease of tracking performance, and eventually lead to the failure of the tracking algorithm. This paper presents a new correlative classifiers approach to address the above problem. Our idea is to derive a group of correlative classifiers based on sample set method. Then we propose strategy to establish the classifiers and to query the suitable classifiers for the next frame tracking. In order to deal with nonlinear problem, particle filter is adopted and integrated with sample set method. For choosing the target from candidate particles, we define a similarity measurement between particles and sample set. The proposed sample set method includes the following steps. First, we cropped positive samples set around the target and negative samples set far away from the target. Second, we extracted average Haar-like feature from these samples and calculate their statistical characteristic which represents the target model. Third, we define the similarity measurement based on the statistical characteristic of these two sets to judge the similarity between candidate particles and target model. Finally, we choose the largest similarity score particle as the target in the new frame. A number of experiments show the robustness and efficiency of the proposed approach when compared with other state-of-the-art trackers.     

RGC: A new conceptual clustering algorithm for mixed incomplete data sets

In this paper, a new conceptual algorithm for the conceptual analysis of mixed incomplete data sets is introduced. This is a logical combinatorial pattern recognition (LCPR) based tool for the conceptual structuralization of spaces. Starting from the limitations of the elaborated conceptual algorithms, our laboratories are working in the application of the methods, the techniques, and in general, the philosophy of the logical combinatorial pattern recognition with the task to improve those limitations. An extension of Michalski's concept of l-complex for any similarity measure, a generalization operator for symbolic variables, and an extension of Michalski's refunion operator are introduced. Finally, the performance of the RGC algorithm is analyzed. A comparison with several known conceptual algorithms is presented.     

Doubling Convex Sets in Lattices: Characterizations and Recognition Algorithms

We characterize lattices obtained from another lattice by a doubling of a convex set. This gives rise to a characterization of the class CN of lattices obtained by doublings of connected and convex sets when starting from a two-element lattice, and from this characterization result we derive an efficient recognition algorithm. This algorithm can be directly applied to the recognition of lattices in the subclasses of CN defined by giving some additionnal constraints on the convex sets used in the doublings.     

An automatic algorithm selection approach for the multi-mode resource-constrained project scheduling problem

This paper investigates the construction of an automatic algorithm selection tool for the multi-mode resource-constrained project scheduling problem (MRCPSP). The research described relies on the notion of empirical hardness models. These models map problem instance features onto the performance of an algorithm. Using such models, the performance of a set of algorithms can be predicted. Based on these predictions, one can automatically select the algorithm that is expected to perform best given the available computing resources. The idea is to combine different algorithms in a super-algorithm that performs better than any of the components individually. We apply this strategy to the classic problem of project scheduling with multiple execution modes. We show that we can indeed significantly improve on the performance of state-of-the-art algorithms when evaluated on a set of unseen instances. This becomes important when lots of instances have to be solved consecutively. Many state-of-the-art algorithms perform very well on a majority of benchmark instances, while performing worse on a smaller set of instances. The performance of one algorithm can be very different on a set of instances while another algorithm sees no difference in performance at all. Knowing in advance, without using scarce computational resources, which algorithm to run on a certain problem instance, can significantly improve the total overall performance.