A Modified Fixed‐Threshold SMO for 1‐Slack Structural SVMs

In this paper, we describe a modified fixed‐threshold sequential minimal optimization (FSMO) for 1‐slack structural support vector machine (SVM) problems. Because the modified FSMO uses the fact that the formulation of 1‐slack structural SVMs has no bias, it breaks down the quadratic programming (QP) problems of 1‐slack structural SVMs into a series of smallest QP problems, each involving only one variable. For various test sets, the modified FSMO is as accurate as existing structural SVM implementations (n‐slack and 1‐slack SVM‐struct) but is faster on large data sets.     

A Lion‐Whale optimization‐based migration of virtual machines for data centers in cloud computing

The scalability, reliability, and flexibility in the cloud computing services are the obligations in the growing demand of computation power. To sustain the scalability, a proper virtual machine migration (VMM) approach is needed with apt balance on quality of service and service‐level agreement violation. In this paper, a novel VMM algorithm based on Lion‐Whale optimization is developed by integrating the Lion optimization algorithm and the Whale optimization algorithm. The optimal virtual machine (VM) migration is performed by the Lion‐Whale VMM based on a new fitness function in the regulation of the resource use, migration cost, and energy consumption of VM placement. The experimentation of the proposed VM migration strategy is performed over 4 cloud setups with a different configuration which are simulated using CloudSim toolkit. The performance of the proposed method is validated over existing optimization‐based VMM algorithms, such as particle swarm optimization and genetic algorithm, using the performance measures, such as energy consumption, migration cost, and resource use. Simulation results reveal the fact that the proposed Lion‐Whale VMM effectively outperforms other existing approaches in optimal VM placement for cloud computing environment with reduced migration cost of 0.01, maximal resource use of 0.36, and minimal energy consumption of 0.09.     

利用密度聚类支持向量机的气象云图云检测

为了提高气象云图云检测的判识精度和计算效率,提出一种基于密度聚类支持向量机(DC-SVM)的云检测方法。分析了MTSAT气象云图的特征提取和选择方案,建立了云和下垫面的分类样本集;在SVM学习中,通过引入样本集的纯度及充足度,选择关键样本,减少了噪声和异常样本的干扰,从而降低了计算复杂度,提高了分类精度。实验表明,该算法的分类正确率较BP神经网络及传统SVM的方法分别提高了2.54%和0.21%,训练时间及测试时间也明显减少;而且,该方法还克服了传统云检测方法需要根据先验知识确定阈值的缺点,检测结果与人工解译结果基本吻合。     

Support vector machine‐based equalisation for direct‐sequence ultra wideband systems

We proposed the support vector machine (SVM)‐based equalisation schemes for direct‐sequence ultra wideband (UWB) systems. The severe intersymbol interference caused by the UWB channel was formulated as a pattern classification problem in the SVM‐based equaliser, which operates in two main modes: training and detection. We also applied the least squares support vector classifiers (LS‐SVCs) to reduce the training complexity and sparse LS‐SVCs to reduce the detection complexity, with little performance loss compared to SVCs. Simulation results confirm the outperformance of the proposed equalisers over the conventional rake receiver with the same order of complexity for detection, especially when no channel information is known at the receiver. Also, the SVM‐based equalisers in the line‐of‐sight scenario provide a performance close to the case with additive white Gaussian noise only. Copyright © 2011 John Wiley & Sons, Ltd.     

Fast Training of Structured SVM Using Fixed‐Threshold Sequential Minimal Optimization

In this paper, we describe a fixed‐threshold sequential minimal optimization (FSMO) for structured SVM problems. FSMO is conceptually simple, easy to implement, and faster than the standard support vector machine (SVM) training algorithms for structured SVM problems. Because FSMO uses the fact that the formulation of structured SVM has no bias (that is, the threshold b is fixed at zero), FSMO breaks down the quadratic programming (QP) problems of structured SVM into a series of smallest QP problems, each involving only one variable. By involving only one variable, FSMO is advantageous in that each QP sub‐problem does not need subset selection. For the various test sets, FSMO is as accurate as an existing structured SVM implementation (SVM‐Struct) but is much faster on large data sets. The training time of FSMO empirically scales between O ( n ) and O(n^1.2), while SVM‐Struct scales between O(n^1.5) and O(n^1.8).     

杂草算法优化支持向量机的网络入侵检测

为了解决支持向量机(优化SVM)在网络入侵检测中的参数优化问题,以提高网络入侵检测性能,提出一种入侵杂草(IWO)算法SVM的网络入侵检测模型(IWO-SVM)。首先将SVM参数编码为入侵杂草,以检测率作为优化目标函数,然后通过模拟杂草入侵种子的生长过程找到最SVM的最优参数,从而最优网络入侵检测模型,后在采用KDD99数据集性能测试。结果表明IWO-SVM是一种检测检测率高、速度快的网络入侵检测模型。     

Imbalanced SVM‐Based Anomaly Detection Algorithm for Imbalanced Training Datasets

Abnormal samples are usually difficult to obtain in production systems, resulting in imbalanced training sample sets. Namely, the number of positive samples is far less than the number of negative samples. Traditional Support Vector Machine (SVM)‐based anomaly detection algorithms perform poorly for highly imbalanced datasets: the learned classification hyperplane skews toward the positive samples, resulting in a high false‐negative rate. This article proposes a new imbalanced SVM (termed ImSVM)‐based anomaly detection algorithm, which assigns a different weight for each positive support vector in the decision function. ImSVM adjusts the learned classification hyperplane to make the decision function achieve a maximum GMean measure value on the dataset. The above problem is converted into an unconstrained optimization problem to search the optimal weight vector. Experiments are carried out on both Cloud datasets and Knowledge Discovery and Data Mining datasets to evaluate ImSVM. Highly imbalanced training sample sets are constructed. The experimental results show that ImSVM outperforms over‐sampling techniques and several existing imbalanced SVM‐based techniques.     

FIPA‐based reference architecture for efficient discovery and selection of appropriate cloud service using cloud ontology

Cloud computing is considered the latest emerging computing paradigm and has brought revolutionary changes in computing technology. With the advancement in this field, the number of cloud users and service providers is increasing continuously with more diversified services. Consequently, the selection of appropriate cloud service has become a difficult task for a new cloud customer. In case of inappropriate selection of a cloud services, a cloud customer may face the vendor locked‐in issue and data portability and interoperability problems. These are the major obstacles in the adoption of cloud services. To avoid these complexities, a cloud customer needs to select an appropriate cloud service at the initial stage of the migration to the cloud. Many researches have been proposed to overcome the issues, but problems still exist in intercommunication standards among clouds and vendor locked‐in issues. This research proposed an IEEE multiagent Foundation for Intelligent Physical Agent (FIPA) compliance multiagent reference architecture for cloud discovery and selection using cloud ontology. The proposed approach will mitigate the prevailing vendor locked‐in issue and also alleviate the portability and interoperability problems in cloud computing. To evaluate the proposed reference architecture and compare it with the state‐of‐the‐art approaches, several experiments have been performed by utilizing the commonly used performance measures. Analysis indicates that the proposed approach enables significant improvements in cloud service discovery and selection in terms of search efficiency, execution, and response time.     

QMLFD Based RSA Cryptosystem for Enhancing Data Security in Public Cloud Storage System

Nowadays sharing secure data turns out to be a challenging task for the data owner due to its privacy and confidentiality. Several IT companies stores their important information in the cloud since computing has developed immense power in sharing the data. On the other hand, privacy is considered a serious issue in cloud computing as there are numerous privacy concerns namely integrity, authentication as well as confidentiality. Among all those concerns, this paper focuses on enhancing the data integrity in the public cloud environment using Qusai modified levy flight distribution for the RSA cryptosystem (QMLFD-RSA). An effective approach named QMLFD for the RSA cryptosystem is proposed for resolving the problem based on data integrity in public cloud environment. A secured key generation and data encryption are done by employing the RSA cryptosystem thus the data is secured from unauthorized users. The key selection is done by using quasi based modified Levy flight distribution algorithm. Thus the proposed approach provides an effective model to enhance the integrity of data in cloud computing thus checking the data integrity uploaded in the public cloud storage system. In addition to this, ten optimization benchmark functions are calculated to determine the performances and the functioning of the newly developed QMLFD algorithm. The simulation results and comparative performances are carried out and the analysis reveals that the proposed QMLFD for the RSA cryptosystem provides better results when compared with other approaches.

     

Beamforming‐based feature extraction and RVM‐based method for attacker node classification in CRN

Cognitive radio is a promising technology for the future wireless spectrum allocation to improve the utilization rate of the licensed bands. However, the cognitive radio network is susceptible to various attacks. Hence, there arises a need to develop a highly efficient security measure against the attacks. This paper presents a beamforming‐based feature extraction and relevance vector machine (RVM)‐based method for the classification of the attacker nodes in the cognitive radio network. Initially, the allocation of the Rayleigh channel is performed for the communication. The quaternary phase shift keying method is used for modulating the signals. After obtaining the modulated signal, the extraction of the beamforming‐based features is performed. The RVM classifier is used for predicting the normal nodes and attacker nodes. If the node is detected as an attacker node, then communication with that node is neglected. Particle swarm optimization is applied for predicting the optimal channel, based on the beamforming feature values. Then, signal communication with the normal nodes is started. Finally, the signal is demodulated. The signal‐to‐noise ratio and bit‐error rate values are computed to evaluate the performance of the proposed approach. The accuracy, sensitivity, and specificity of the RVM classifier method are higher than the support vector machine classifier. The proposed method achieves better performance in terms of throughput, channel sensing/probing rate, and channel access delay. Copyright © 2016 John Wiley & Sons, Ltd.     

支持向量机中的模型选择研究

支持向量机是一种新型的机器学习方法。模型选择是设计支持向量机的关键。本文在分析用于分类的支持向量机原理的基础上,分别从核函数类型和核参数的选择等模型选择方面进行了探讨。最后在上述理论分析的基础上进行了实验,取得了较好的效果。     

Condition multi-classification and evaluation of system degradation process using an improved support vector machine

Degradation process is a non-negligible phenomenon in system condition monitoring and reliability practices. Traditional binary-state characterization (i.e., normal and failure) on system health condition may not provide accurate information for maintenance scheduling, and the multi-state classification for degradation process is a necessary step to realize cost-effective condition based maintenance. Support vector machine (SVM) is a useful technique for system condition monitoring and fault diagnosis. However, the SVM classification accuracy of deteriorating system is usually poor, because characteristics of different degradation states may not be very distinctive. This paper presented an improved support vector machine for system degradation classification and evaluation. The core of the proposed method can be summarized as: an improved voting scheme in one-against-one SVM, and an optimization of classification process by utilizing inherent physical property of system state transition. A case study of cooling fan bearing accelerated life time test is conducted to obtain the experimental data, and a comparison before and after optimization shows that the proposed method improves the classification accuracy.     

A secure cost‐effective migration of enterprise applications to the cloud

Cloud computing is rapidly expanding as an alternative service deployment platform today. This brings forth many new challenges in migrating enterprise applications into cloud. To enable enterprises to benefit from migration while achieving cost‐efficiency and keeping sensitive user data confidential against untrusted servers, planning which servers to migrate to the cloud and which to be hosted on‐premise is a key problem. This problem has been traditionally approached through the formulation and resolution of large optimization problems requiring global knowledge. Such approaches are not suitable for large‐scale and dynamic enterprise network migrations. In this paper, the problem of determining the optimal migrated components set of an enterprise application is revisited and addressed in a way that is both scalable and deals inherently with network dynamicity. In particular, application migration, which enables service components to move between local data center toward more communication cost‐effective cloud, is based on local information. The migration policies proposed in this work are analytically shown to be capable of moving an enterprise application between local data center and remote cloud in a way that the cost of service provision is reduced. Experimental results show the efficiency, applicability, and easy adaptability of the presented approach.Copyright © 2013 John Wiley & Sons, Ltd.     

Support vector machine-based classification scheme for myoelectric control applied to upper limb

This paper proposes and evaluates the application of support vector machine (SVM) to classify upper limb motions using myoelectric signals. It explores the optimum configuration of SVM-based myoelectric control, by suggesting an advantageous data segmentation technique, feature set, model selection approach for SVM, and postprocessing methods. This work presents a method to adjust SVM parameters before classification, and examines overlapped segmentation and majority voting as two techniques to improve controller performance. A SVM, as the core of classification in myoelectric control, is compared with two commonly used classifiers: linear discriminant analysis (LDA) and multilayer perceptron (MLP) neural networks. It demonstrates exceptional accuracy, robust performance, and low computational load. The entropy of the output of the classifier is also examined as an online index to evaluate the correctness of classification; this can be used by online training for long-term myoelectric control operations.     

An accurate traffic classification model based on support vector machines

Network traffic classification is a fundamental research topic on high‐performance network protocol design and network operation management. Compared with other state‐of‐the‐art studies done on the network traffic classification, machine learning (ML) methods are more flexible and intelligent, which can automatically search for and describe useful structural patterns in a supplied traffic dataset. As a typical ML method, support vector machines (SVMs) based on statistical theory has high classification accuracy and stability. However, the performance of SVM classifier can be severely affected by the data scale, feature dimension, and parameters of the classifier. In this paper, a real‐time accurate SVM training model named SPP‐SVM is proposed. An SPP‐SVM is deducted from the scaling dataset and employs principal component analysis (PCA) to extract data features and verify its relevant traffic features obtained from PCA. By employing PCA algorithm to do the dimension extraction, SPP‐SVM confirms the critical component features, reduces the redundancy among them, and lowers the original feature dimension so as to reduce the over fitting and increase its generalization effectively. The optimal working parameters of kernel function used in SPP‐SVM are derived automatically from improved particle swarm optimization algorithm, which will optimize the global solution and make its inertia weight coefficient adaptive without searching for the parameters in a wide range, traversing all the parameter points in the grid and adjusting steps gradually. The performance of its two‐ and multi‐class classifiers is proved over 2 sets of traffic traces, coming from different topological points on the Internet. Experiments show that the SPP‐SVM's two‐ and multi‐class classifiers are superior to the typical supervised ML algorithms and performs significantly better than traditional SVM in classification accuracy, dimension, and elapsed time.     

基于PSO-GWO-SVM的周界安防信号识别研究

针对支持向量机(SVM)在大规模入侵信号分类时存在的局限性,提出了一种改进的SVM信号识别方法。该方法首先采用粒子群优化算法(PSO)来生成多样化的初始位置,然后利用灰狼优化算法(GWO)更新离散搜索空间中样本的当前位置,获得最优特征子集;最后基于最优特征子集用SVM对待测样本进行分类识别。实验结果显明,在识别周界入侵信号时,基于PSO-GWO-SVM算法的分类器获得了96.86%的准确率、95.82%的灵敏度(SE)和96.31%的特异性。与传统的信号识别方法相比,具有更优异的识别精度、适应性和时效性。     

Support Vector Machine Learning for Region‐Based Image Retrieval with Relevance Feedback

We present a relevance feedback approach based on multi‐class support vector machine (SVM) learning and cluster‐merging which can significantly improve the retrieval performance in region‐based image retrieval. Semantically relevant images may exhibit various visual characteristics and may be scattered in several classes in the feature space due to the semantic gap between low‐level features and high‐level semantics in the user's mind. To find the semantic classes through relevance feedback, the proposed method reduces the burden of completely re‐clustering the classes at iterations and classifies multiple classes. Experimental results show that the proposed method is more effective and efficient than the two‐class SVM and multi‐class relevance feedback methods.     

Classification‐Based Approach for Hybridizing Statistical and Rule‐Based Machine Translation

In this paper, we propose a classification‐based approach for hybridizing statistical machine translation and rule‐based machine translation. Both the training dataset used in the learning of our proposed classifier and our feature extraction method affect the hybridization quality. To create one such training dataset, a previous approach used auto‐evaluation metrics to determine from a set of component machine translation (MT) systems which gave the more accurate translation (by a comparative method). Once this had been determined, the most accurate translation was then labelled in such a way so as to indicate the MT system from which it came. In this previous approach, when the metric evaluation scores were low, there existed a high level of uncertainty as to which of the component MT systems was actually producing the better translation. To relax such uncertainty or error in classification, we propose an alternative approach to such labeling; that is, a cut‐off method. In our experiments, using the aforementioned cut‐off method in our proposed classifier, we managed to achieve a translation accuracy of 81.5% — a 5.0% improvement over existing methods.     

基于人工蜂群算法的支持向量机参数优化及应用

为了解决常用的支持向量机(SVM)参数优化方法在寻优过程不同程度的陷入局部最优解的问题,提出一种基于人工蜂群(ABC)算法的SVM参数优化方法。将SVM的惩罚因子和核函数参数作为食物源位置,分类正确率作为适应度,利用ABC算法寻找适应度最高的食物源位置。利用4个标准数据集,将其与遗传(GA)算法、蚁群(ACO)算法、标准粒子群(PSO)算法优化的SVM进行性能比较,结果表明,本文方法能克服局部最优解,获得更高的分类正确率,并在小数目分类问题上有效降低运行时间。将本文方法运用到计算机笔迹鉴别,对提取的笔迹特征进行分类,与GA算法、ACO算法、PSO算法优化的SVM相比,得到了更高的分类正确率。     

Enhanced security‐aware technique and ontology data access control in cloud computing

Nowadays, security and data access control are some of the major concerns in the cloud storage unit, especially in the medical field. Therefore, a security‐aware mechanism and ontology‐based data access control (SA‐ODAC) has been developed to improve security and access control in cloud computing. The model proposed in this research work is based on two operational methods, namely, secure awareness technique (SAT) and ontology‐based data access control (ODAC), to improve security and data access control in cloud computing. The SAT technique is developed to provide security for medical data in cloud computing, based on encryption, splitting and adding files, and decryption. The ODAC ontology is launched to control unauthorized persons accessing data from storage and create owner and administrator rules to allow access to data and is proposed to improve security and restrict access to data. To manage the key of the SAT technique, the secret sharing scheme is introduced in the proposed framework. The implementation of the algorithm is performed by MATLAB, and its performance is verified in terms of delay, encryption time, encryption time, and ontology processing time and is compared with role‐based access control (RBAC), context‐aware RBAC and context‐aware task RBAC, and security analysis of advanced encryption standard and data encryption standard. Ultimately, the proposed data access control and security scheme in SA‐ODAC have achieved better performance and outperform the conventional technique.