基于特征选择的VNF资源需求预测方法

针对在网络切片场景下以往的VNF(虚拟网络功能)资源分配策略无法满足动态的资源需求,很容易导致资源分配不足或过度分配的问题,提出了一种基于两阶段算法(two-stage algorithm,TSA)的VNF资源需求预测方法.该方法首先基于数据特征筛选出与预测目标高度相关的候选特征集,然后利用贪婪式前向搜索策略对候选特征集进一步筛选获得最优特征集,最终训练出不同类型的预测模型.仿真结果表明,基于该方法所训练的模型可以获得更好的预测性能,同时该方法的可扩展性较好,训练好的模型可以直接集成到现有的VNF部署算法中应用.     

基于资源聚集的计算网格备份资源选择算法

资源备份是提高计算网格应用可用性的重要方法．如何为应用选择备份资源是网格资源备份服务要解决的首要问题．文章提出了基于资源聚集的备份资源选择算法．该算法将为应用分配的资源按照资源之间的关系聚集成多个资源集合，然后根据应用的可用性需求为每个资源集合选择备份资源，每个资源集合中的资源共享同一组备份资源．这一算法适用于计算网格，可以在资源备份服务模块中实现．最后，给出了该算法的应用实例．     

基因数据集混合特征选择算法研究

基因数据的特点是高维度、小样本、大噪声,在处理过程中容易造成维数灾难和过度拟合等问题。针对这种情况提出一种新的基因数据集的特征选择方法,第一步是通过ReliefF算法对基因特征进行权重重要度的筛选;第二步是对筛选过的特征集合进行mRMR算法判断,留下与目标类别高度相关而其间相关性较小的基因特征;第三步利用邻域粗糙集特征选择算法对简化后的基因数据集进行寻优处理,选出最优化的特征基因子集。为了证明新算法的有效性,以SVM为分类器,使用外部交叉验证法对整个过程来计算,从而验证本文新特征选择方法的有效性。     

An efficient accelerator for attribute reduction from incomplete data in rough set framework

Feature selection (attribute reduction) from large-scale incomplete data is a challenging problem in areas such as pattern recognition, machine learning and data mining. In rough set theory, feature selection from incomplete data aims to retain the discriminatory power of original features. To address this issue, many feature selection algorithms have been proposed, however, these algorithms are often computationally time-consuming. To overcome this shortcoming, we introduce in this paper a theoretic framework based on rough set theory, which is called positive approximation and can be used to accelerate a heuristic process for feature selection from incomplete data. As an application of the proposed accelerator, a general feature selection algorithm is designed. By integrating the accelerator into a heuristic algorithm, we obtain several modified representative heuristic feature selection algorithms in rough set theory. Experiments show that these modified algorithms outperform their original counterparts. It is worth noting that the performance of the modified algorithms becomes more visible when dealing with larger data sets.     

Impact of error estimation on feature selection

Given a large set of potential features, it is usually necessary to find a small subset with which to classify. The task of finding an optimal feature set is inherently combinatoric and therefore suboptimal algorithms are typically used to find feature sets. If feature selection is based directly on classification error, then a feature-selection algorithm must base its decision on error estimates. This paper addresses the impact of error estimation on feature selection using two performance measures: comparison of the true error of the optimal feature set with the true error of the feature set found by a feature-selection algorithm, and the number of features among the truly optimal feature set that appear in the feature set found by the algorithm. The study considers seven error estimators applied to three standard suboptimal feature-selection algorithms and exhaustive search, and it considers three different feature-label model distributions. It draws two conclusions for the cases considered: (1) depending on the sample size and the classification rule, feature-selection algorithms can produce feature sets whose corresponding classifiers possess errors far in excess of the classifier corresponding to the optimal feature set; and (2) for small samples, differences in performances among the feature-selection algorithms are less significant than performance differences among the error estimators used to implement the algorithms. Moreover, keeping in mind that results depend on the particular classifier-distribution pair, for the error estimators considered in this study, bootstrap and bolstered resubstitution usually outperform cross-validation, and bolstered resubstitution usually performs as well as or better than bootstrap.     

Formulation of radiometric feasibility measures for feature selection and planning in visual servoing

Feature selection and planning are integral parts of visual servoing systems. Because many irrelevant and nonreliable image features usually exist, higher accuracy and robustness can be expected by selecting and planning good features. Assumption of perfect radiometric conditions is common in visual servoing. The following paper discusses the issue of radiometric constraints for feature selection in the context of visual servoing. Here, radiometric constraints are presented and measures are formulated to select the optimal features (in a radiometric sense) from a set of candidate features. Simulation and experimental results verify the effectiveness of the proposed measures.     

Selecting salient features for classification based on neural network committees

Aggregating outputs of multiple classifiers into a committee decision is one of the most important techniques for improving classification accuracy. The issue of selecting an optimal subset of relevant features plays also an important role in successful design of a pattern recognition system. In this paper, we present a neural network based approach for identifying salient features for classification in neural network committees. Feature selection is based on two criteria, namely the reaction of the cross-validation data set classification error due to the removal of the individual features and the diversity of neural networks comprising the committee. The algorithm developed removed a large number of features from the original data sets without reducing the classification accuracy of the committees. The accuracy of the committees utilizing the reduced feature sets was higher than those exploiting all the original features.     

Robust multi-view feature matching from multiple unordered views

This paper explores the problem of multi-view feature matching from an unordered set of widely separated views. A set of local invariant features is extracted independently from each view. First we propose a new view-ordering algorithm that organizes all the unordered views into clusters of related (i.e. the same scene) views by efficiently computing the view-similarity values of all view pairs by reasonably selecting part of extracted features to match. Second a robust two-view matching algorithm is developed to find initial matches, then detect the outliers and finally incrementally find more reliable feature matches under the epipolar constraint between two views from dense to sparse based on an assumption that changes of both motion and feature characteristics of one match are consistent with those of neighbors. Third we establish the reliable multi-view matches across related views by reconstructing missing matches in a neighboring triple of views and efficiently determining the states of matches between view pairs. Finally, the reliable multi-view matches thus obtained are used to automatically track all the views by using a self-calibration method. The proposed methods were tested on several sets of real images. Experimental results show that it is efficient and can track a large set of multi-view feature matches across multiple widely separated views.     

Differential evolution algorithm as a tool for optimal feature subset selection in motor imagery EEG

One of the challenges in developing a Brain Computer Interface (BCI) is dealing with the high dimensionality of the data when extracting features from EEG signals. Different feature selection algorithms have been proposed to overcome this problem but most of them involve complex transformed features, which require high computation and also result in increasing size of the feature set. In this paper, we present a new hybrid method to select features that involves a Differential Evolution (DE) optimization algorithm for searching the feature space to generate the optimal feature subset, with performance evaluated by a classifier. We provide a comprehensive study of the significance of evolutionary algorithm in selecting the best features for EEG signals. The BCI competition III, dataset IVa has been used to evaluate the method. Experimental results demonstrate that the proposed method performs well with Support Vector Machine (SVM) classifier, with an average classification accuracy of above 95% with a minimum of just 10 features. We also present a comparison of Differential Evolution (DE) with other evolutionary algorithms, and the results show the superiority of DE which implies that, with the selection of a good searching algorithm, a simple Common Spatial Pattern filter features can produce good results.     

An eye–hand data fusion framework for pervasive sensing of surgical activities

This paper describes a generic framework for activity recognition based on temporal signals acquired from multiple input modalities and demonstrates its use for eye–hand data fusion. As a part of the data fusion framework, we present a multi-objective Bayesian Framework for Feature Selection with a pruned-tree search algorithm for finding the optimal feature set(s) in a computationally efficient manner. Experiments on endoscopic surgical episode recognition are used to investigate the potential of using eye-tracking for pervasive monitoring of surgical operation and to demonstrate how additional information induced by hand motion can further enhance the recognition accuracy. With the proposed multi-objective BFFS algorithm, suitable feature sets both in terms of feature relevancy and redundancy can be identified with a minimal number of instruments being tracked.     

Using Rough Sets with Heuristics for Feature Selection

Practical machine learning algorithms are known to degrade in performance (prediction accuracy) when faced with many features (sometimes attribute is used instead of feature) that are not necessary for rule discovery. To cope with this problem, many methods for selecting a subset of features have been proposed. Among such methods, the filter approach that selects a feature subset using a preprocessing step, and the wrapper approach that selects an optimal feature subset from the space of possible subsets of features using the induction algorithm itself as a part of the evaluation function, are two typical ones. Although the filter approach is a faster one, it has some blindness and the performance of induction is not considered. On the other hand, the optimal feature subsets can be obtained by using the wrapper approach, but it is not easy to use because of the complexity of time and space. In this paper, we propose an algorithm which is using rough set theory with greedy heuristics for feature selection. Selecting features is similar to the filter approach, but the evaluation criterion is related to the performance of induction. That is, we select the features that do not damage the performance of induction.     

Feature selection based on improved ant colony optimization for online detection of foreign fiber in cotton

Feature selection plays an important role in the machine-vision-based online detection of foreign fibers in cotton because of improvement detection accuracy and speed. Feature sets of foreign fibers in cotton belong to multi-character feature sets. That means the high-quality feature sets of foreign fibers in cotton consist of three classes of features which are respectively the color, texture and shape features. The multi-character feature sets naturally contain a space constraint which lead to the smaller feature space than the general feature set with the same number of features, however the existing algorithms do not consider the space characteristic of multi-character feature sets and treat the multi-character feature sets as the general feature sets. This paper proposed an improved ant colony optimization for feature selection, whose objective is to find the (near) optimal subsets in multi-character feature sets. In the proposed algorithm, group constraint is adopted to limit subset constructing process and probability transition for reducing the effect of invalid subsets and improve the convergence efficiency. As a result, the algorithm can effectively find the high-quality subsets in the feature space of multi-character feature sets. The proposed algorithm is tested in the datasets of foreign fibers in cotton and comparisons with other methods are also made. The experimental results show that the proposed algorithm can find the high-quality subsets with smaller size and high classification accuracy. This is very important to improve performance of online detection systems of foreign fibers in cotton.     

A feature cluster taxonomy based feature selection technique

Feature subset selection is basically an optimization problem for choosing the most important features from various alternatives in order to facilitate classification or mining problems. Though lots of algorithms have been developed so far, none is considered to be the best for all situations and researchers are still trying to come up with better solutions. In this work, a flexible and user-guided feature subset selection algorithm, named as FCTFS (Feature Cluster Taxonomy based Feature Selection) has been proposed for selecting suitable feature subset from a large feature set. The proposed algorithm falls under the genre of clustering based feature selection techniques in which features are initially clustered according to their intrinsic characteristics following the filter approach. In the second step the most suitable feature is selected from each cluster to form the final subset following a wrapper approach. The two stage hybrid process lowers the computational cost of subset selection, especially for large feature data sets. One of the main novelty of the proposed approach lies in the process of determining optimal number of feature clusters. Unlike currently available methods, which mostly employ a trial and error approach, the proposed method characterises and quantifies the feature clusters according to the quality of the features inside the clusters and defines a taxonomy of the feature clusters. The selection of individual features from a feature cluster can be done judiciously considering both the relevancy and redundancy according to user’s intention and requirement. The algorithm has been verified by simulation experiments with different bench mark data set containing features ranging from 10 to more than 800 and compared with other currently used feature selection algorithms. The simulation results prove the superiority of our proposal in terms of model performance, flexibility of use in practical problems and extendibility to large feature sets. Though the current proposal is verified in the domain of unsupervised classification, it can be easily used in case of supervised classification.     

On setup level tool sequence selection for 2.5-D pocket machining

This paper describes algorithms for efficiently machining an entire setup. Previously, the author developed a graph based algorithm to find the optimal tool sequence for machining a single 2.5-axis pocket. This paper extends this algorithm for finding an efficient tool sequence to machine an entire setup. A setup consists of a set of features with precedence constraints, that are machined when the stock is clamped in a particular orientation. The precedence constraints between the features primarily result from nesting of some features within others. Four extensions to the basic graph algorithm are investigated in this research. The first method finds optimal tool sequences on a feature by feature basis. This is a local optimization method that does not consider inter feature tool-path interactions. The second method uses a composite graph for finding an efficient tool sequence for the entire setup. The constrained graph and subgraph approaches have been developed for situations where different features in the setup have distinct critical tools. It is found that the first two methods can produce erroneous results which can lead to machine crashes and incomplete machining. Illustrative examples have been generated for each method.     

Comparative study for feature selection algorithms in intrusion detection system

The Intrusion Detection System (IDS) deals with the huge amount of network data that includes redundant and irrelevant features causing slow training and testing procedure, higher resource usage and poor detection ratio. Feature selection is a vital preprocessing step in intrusion detection. Hence, feature selec-tion is an essential issue in intrusion detection and need to be addressed by selec-ting the appropriate feature selection algorithm. A major challenge to select the optimal feature selection methods can precisely calculate the relevance of fea-tures to the detection process and the redundancy among features. In this paper, we study the concepts and algorithms used for feature selection algorithms in the IDS. We conclude this paper by identifying the best feature selection algorithm to select the important and useful features from the network dataset.     

基于相容关系的基因选择方法

有效的基因选择是对基因表达数据进行分析的重要内容。粗糙集作为一种软计算方法能够保持在数据集分类能力不变的基础上,对属性进行约简。由于基因表达数据的连续性,为了避免运用粗糙集方法所必需的离散化过程带来的信息丢失,将相容粗糙集应用于基因的特征选取,提出了基于相容关系的基因选择方法。首先,通过i检验对基因表达数据进行排列,选择评分靠前的若干基因;然后,通过相容粗糙集对这些基因进一步约简。在两个标准的基因表达数据上进行了实验,结果表明该方法是可行性和有效性的。     

小样本情况下Fisher线性鉴别分析的理论及其验证

线性鉴别分析是特征抽取中最为经典和广泛使用的方法之一。近几年,在小样本情况下如何抽取F isher最优鉴别特征一直是许多研究者关心的问题。本文应用投影变换和同构变换的原理,从理论上解决了小样本情况下最优鉴别矢量的求解问题,即最优鉴别矢量可在一个低维空间里求得;给出了特征抽取模型,并给出求解模型的PPCA+LDA算法;在ORL人脸库3种分辨率灰度图像上进行实验。实验结果表明,PPCA+LDA算法抽取的鉴别向量有较强的特征抽取能力,在普通的最小距离分类器下能达到较高的正确识别率,而且识别结果十分稳定。     

Process sequencing and process clustering in process planning using state space search

Process sequencing, as a very important part of process planning, has been the subject of many research reports in the area of process planning, but is usually treated as a feature-sequencing problem. This paper presents a novel algorithm for process sequencing, which considers the feature precedence network, different process candidates, and machine and tool constraints. The algorithm consists of two parts: process clustering and process sequencing. For clustering we used a notion of the same resource usage for different features, while for sequencing we applied the best-first search method algorithm to generate an optimal process sequence. The algorithm has been applied on several examples with realistic complexities, and it showed satisfactory results.     

Optimal control for integrated emission management in diesel engines

Integrated Emission Management (IEM) is a supervisory control strategy that minimises operational costs (consisting of fuel and AdBlue) for diesel engines with an aftertreatment system, while satisfying emission constraints imposed by legislation. In most work on IEM, a suboptimal heuristic real-time implementable solution is used, which is based on Pontryagin's Minimum Principle (PMP). In this paper, we compute the optimal solution using both PMP and Dynamic Programming (DP). As the emission legislation imposes a terminal state constraint, standard DP algorithms are sensitive to numerical errors that appear close to the boundary of the feasible sets. Therefore, we propose two extensions to existing DP methods, which use an approximation of the forward reachable sets to reduce the grid size over time and an approximation of the backward reachable sets to avoid the aforementioned numerical errors. Using a simulation study of a cold-start World Harmonised Transient Cycle for a Euro-VI engine, we show that the novel extension to the DP algorithm yields the best approximation of the optimal cost, when compared to existing DP methods. Furthermore, we show that PMP yields almost the same results as DP, and that the real-time implementable solution only deviates approximately 0.08–0.16% from the optimal solution.     

Wavelet kernel learning

This paper addresses the problem of optimal feature extraction from a wavelet representation. Our work aims at building features by selecting wavelet coefficients resulting from signal or image decomposition on an adapted wavelet basis. For this purpose, we jointly learn in a kernelized large-margin context the wavelet shape as well as the appropriate scale and translation of the wavelets, hence the name “wavelet kernel learning”. This problem is posed as a multiple kernel learning problem, where the number of kernels can be very large. For solving such a problem, we introduce a novel multiple kernel learning algorithm based on active constraints methods. We furthermore propose some variants of this algorithm that can produce approximate solutions more efficiently. Empirical analysis show that our active constraint MKL algorithm achieves state-of-the art efficiency. When used for wavelet kernel learning, our experimental results show that the approaches we propose are competitive with respect to the state-of-the-art on brain–computer interface and Brodatz texture datasets.