多分类器融合在语音情感识别中的应用

针对语音情感识别问题,提出一种采用决策模板的多分类器融合方法,利用不同类型的声学特征子集来构造子分类器。不同的子集能充分提高各子分类器之间的“多样性”指标,这是多分类器融合算法能够成功应用的必备条件。与多数投票融合算法和支持向量机相比该方法取得了较好的识别结果。另一方面,从多样性指标分析的角度出发探究该方法能获得较好识别效果的原因。     

基于稀疏表示和决策融合的图像分类方法

利用多个稀疏表示分类器融合的决策信息对图像进行分类,可避免单个特征对图像分类的影响。提出一种自适应调节权重的多稀疏分类器融合图像分类方法。对原始图像分别提取3组不同特征,并训练出各自稀疏表示分类器;根据各个子分类器的准确率,通过迭代计算自适应确定各分类器最终权重;融合各子分类器的输出结果进行最终类别判断。基于Cifar-10图像数据集进行多组实验,结果表明,相对仅提取单特征的图像分类方法,该方法有效提高了图像分类准确率。     

双色红外图像目标多特征决策级融合识别算法

针对双色红外成像系统中的自动目标识别问题,提出了一种采用多特征多分类器决策级融合的目标识别算法。该算法首先提取目标的形状特征和面貌特征;接着基于各种不同特征设计多个分类器对目标进行分类;然后采用所设计的多分类器决策级融合策略对多个分类器的目标分类结果进行融合处理;最后采用所提出的决策规则对多分类器融合分类结果进行处理得到最终的目标识别结果。该算法充分利用了目标在多传感器图像中的多种分类特征信息,在较大程度上提高了系统的目标识别效率和精确性。实验结果证实了该算法的有效性。     

一种新的启发式分类器选择方法

分类器选择是一种设计多分类器系统的有效方法,从给定候选分类器集中挑选出一个子集,使得该子集集成性能最佳。现有的分类器选择方法大多采用基于集成精度的随机搜索方法,但巨大的搜索复杂度限制了它们在更大系统中的应用。该文提出一种新的选择标准——IWCECR及一种基于IWCECR的启发式搜索算法,在手写体数字识别的实验中,从20个候选分类器中挑选子集,结果表明,该方法具有较高的搜索效率,在子集集成性能方面仅次于穷举法。     

半动态集成选择分类方法

在集成学习领域,传统的动态集成选择需要为每一个样本选择子分类器组成集成分类器,这极大地增加了计算复杂度。针对这一问题,提出一种新的半动态集成选择方法。该方法分为两阶段,第一阶段为所有的测试样本选择最好的个体分类器组成一个集成分类器,第二阶段从剩余的个体分类器集合中为当前测试样本动态地选择子分类器组成一个集成分类器。最终的分类结果通过融合两阶段得到集成分类器的结果得到。通过对UCI数据测试的结果表明,该算法不仅能取得较好的分类性能,而且能极大地降低计算复杂度。     

多分类器选择集成方法

针对目前人们对分类性能的高要求和多分类器集成实现的复杂性,从基分类器准确率和基分类器间差异性两方面出发,提出了一种新的多分类器选择集成算法。该算法首先从生成的基分类器中选择出分类准确率较高的,然后利用分类器差异性度量来选择差异性大的高性能基分类器,在分类器集成之前先对分类器集进行选择获得新的分类器集。在UCI数据库上的实验结果证明,该方法优于bagging方法,取得了很好的分类识别效果。     

基于动态权值集成的手写数字识别方法

提出了一种基于动态权值集成的多分类器手写数字识别方法。该方法采用BP神经网络的方法,对不同的特征输入向量构建不同的神经网络分类器,通过设定动态权值,进而对不同的分类器的输出向量采用多类器集成方法进行系统集成。实验结果表明该方法具有较高的识别率和识别精度。     

基于SVM的多分类器融合算法在说话人识别中的应用

多分类器协同合作克服了单个分类器识别效果不理想,适用范围较窄或对使用环境要求较高的不足,并为研制更高性能的分类器,提供了解决问题的另一个途径.提出了一种基于SVM的多分类器说话人识别系统,该系统的识别框架基于多分类器的协同工作.该多分类器系统采用了ANN(artificial neural networks)、GMM(gaussian mixed model)分类器和子带结构分类器,参数选取包括MFCC(mel frequency cepstrum coefficient)、LPCC(linear prediction cepstrum coefficient).多分类器融合采用SVM方法.本系统在超短波信道(15 db)的实际应用中达到94%的识别率.     

基于遗传模糊分类的步态识别

提出基于遗传模糊分类的步态识别方法.采用新的特征提取方法,该方法将目标按人体结构特点划分为多个子区域,利用各个子区域的质心与头部质心形成的距离和夹角对步态特征进行描述.运用模糊聚类算法构建不同距离函数的分类器,并用遗传算法对分类器进行集成,组成的集成融合分类器对步态序列进行识别.实验结果表明该方法具有较高的识别性能.     

基于精简随机子空间的多生成树集成一类分类算法

由于高维数据通常存在冗余和噪声,在其上直接构造覆盖模型不能充分反映数据的分布信息,导致分类器性能下降.为此提出一种基于精简随机子空间多树集成分类方法.该方法首先生成多个随机子空间,并在每个子空间上构造独立的最小生成树覆盖模型.其次对每个子空间上构造的分类模型进行精简处理,通过一个评估准则(AUC值),对生成的一类分类器进行精简.最后均值合并融合这些分类器为一个集成分类器.实验结果表明,与其它直接覆盖分类模型和bagging算法相比,多树集成覆盖分类器具有更高的分类正确率.     

云南香格里拉地区森林优势树种决策融合分类

基于Google Earth Engine（GEE）云计算平台,协同Sentinel-2影像、WordClim生物气候数据、SRTM地形数据、森林资源二类调查数据等数据,以随机森林（Random Forest, RF）,支持向量机（Support Vector Machine, SVM）和最大熵（Maximum Entropy, MaxEnt）3种机器学习算法为组件分类器,开展多源特征、多分类器决策融合的优势树种分类研究。通过3种组件分类器分别构建了两种串行集成和3种贝叶斯并行集成模型,用于确定云南香格里拉地区10种主要优势树种的空间分布。分类结果显示：3个组件分类器的总体精度均低于67.17%;3种并行集成方法总体精度相当,约为72%;两种串行集成方法精度高于78.48%,其中MaxEnt-SVM串行集成方法获得最佳精度（OA：80.66%, Kappa：0.78）,与组件分类器相比精度至少提高了13.49%。研究表明：决策融合方法在优势树种分类中比组件分类器精度更高,并且有效改善了小样本树种的分类精度,可用于大范围山区优势树种分类。     

基于子模式的Gabor特征融合的单样本人脸识别

针对传统人脸识别方法在单训练样本条件下效果不佳的缺点,提出基于子模式的Gabor特征融合方法并用于单样本人脸识别。首先采用Gabor变换抽取人脸局部信息,为有效利用面部器官的空间位置信息,将Gabor人脸图像分块构成子模式,采用最小距离分类器对各子模式分类。最后对各子模式分类结果做决策级融合得出分类结果。根据子模式构成原则和决策级融合策略不同,提出两种子模式Gabor特征融合方法。利用ORL人脸库和CAS-PEAL-R1人脸库进行实验和比较分析,实验结果表明文中方法有效提高单样本人脸识别的正确率,改善单样本人脸识别系统的性能。     

A highly modular and extensible architecture for an integrated IMS‐based authoring system: the <e‐Aula> experience

The <e‐Aula> platform is a new experimental e‐learning environment that adheres closely to IMS Global Learning Consortium, Inc. e‐learning standards in order to facilitate their applicability in different learning scenarios. <e‐Aula> is equipped with an integrated modular and extensible architecture for the authoring of IMS‐compliant learning materials focused on the IMS manifest. This manifest‐driven architecture facilitates maintenance and promotes the evolution of the authoring system in <e‐Aula>, both of which are mandatory requirements in the successful production and maintenance of content for many different specialized learning domains. In this paper, we describe this authoring system, its manifest‐driven architecture and its implementation using well‐known and robust Java‐based Web technologies. Copyright © 2006 John Wiley & Sons, Ltd.     

多类型分类器融合的文本分类方法研究

传统的文本分类方法大多数使用单一的分类器,而不同的分类器对分类任务的侧重点不同,就使得单一的分类方法有一定的局限性,同时每个特征提取方法对特征词的考虑角度不同。针对以上问题,提出了多类型分类器融合的文本分类方法。该模型使用了word2vec、主成分分析、潜在语义索引以及TFIDF特征提取方法作为多类型分类器融合的特征提取方法。并在多类型分类器加权投票方法中忽略了类别信息的问题,提出了类别加权的分类器权重计算方法。通过实验结果表明,多类型分类器融合方法在二元语料库、多元语料库以及特定语料库上都取得了很好的性能,类别加权的分类器权重计算方法比多类型分类器融合方法在分类性能方面提高了1.19%。     

Adaptive Fusion of Speech and Lip Information for Robust Speaker Identification

Wark, T., and Sridharan, S., Adaptive Fusion of Speech and Lip Information for Robust Speaker Identification, Digital Signal Processing11 (2001) 169–186This paper compares techniques for asynchronous fusion of speech and lip information for robust speaker identification. In any fusion system, the ultimate challenge is to determine the optimal way to combine all information sources under varying conditions. We propose a new method for estimating confidence levels to allow intelligent fusion of the audio and visual data. We describe a secondary classification system, where secondary classifiers are used to give approximations for the estimation errors of outputs likelihoods from primary classifiers. The error estimates are combined with a dispersion measure technique allowing an adaptive fusion strategy based on the level of data degradation at the time of testing. We compare the performance of this fusion system with two other approaches to linear fusion and show that the use of secondary classifiers is an effective technique for improving classification performance. Identification experiments are performed on the M2VTS multimodal database , with encouraging results.     

A class decomposition approach for GA-based classifiers

Genetic algorithm (GA) has been used as a conventional method for classifiers to evolve solutions adaptively for classification problems. In this paper, a new approach using class decomposition is proposed to improve the performance of GA-based classifiers. A classification problem is fully partitioned into several class modules in the output domain and each module is responsible for solving a fraction of the original problem. These modules are trained in parallel and independently and the results obtained are integrated and evolved further for a final solution. A scheme based on Fisher's linear discriminant (FLD) computation is used to estimate the difficulty of separating two classes. Based on the FLD information derived, different integration approaches are proposed and their performance is compared. The experiment results on a benchmark data set show that class decomposition can achieve higher classification rate than the normal GA and FLD-based integration improves the classification accuracy further.     

一种双目标排序层分类器融合方法

最近 Melnik 提出了一种新的排序层分类器融合思想, 指出在分类器融合过程中既要调节对不同分类器的侧重程度, 又要利用不同序号值提供的置信度信息. 但是在 Melnik 提出的融合方法中, 参数数量随着分类器数量的增加呈指数级增长, 在分类器数目增加时会产生维数灾难问题. 在 Melnik 的思想启发下, 本文提出了一种新的融合方法, 该方法将对序号的变换与分类器的加权组合协调起来, 能够更好地实现 Melnik 提出的目标. 另外, 本文给出了一种用连续可微函数表示的分类错误率表达式, 设计了基于梯度下降的参数调节方法. 在实验中本文设计了融合掌纹图像数据和手指图像数据的多模态身份识别系统, 观察了不同数目分类器条件下的融合效果. 实验结果表明本文方法的分类正确率高于传统方法和 Melnik 的方法.     

Central-tendency estimation and nearest-estimate classification of event related potentials

Event related potentials (ERPs) are modeled as random vectors in order to determine multivariate central-tendency (C-T) estimates of ERPs such as the arithmetic mean, geometric mean, harmonic mean, median, tri-mean, trimmed-mean, and the Winsorized mean. Additionally, it is shown that the C-T estimates can be used to implement various forms of minimum-distance classifiers for individual channels and for single-channel heterogeneous, multi-channel homogeneous, and multi-channel heterogeneous-homogenous ERP classification through decision fusion. The study also focuses on answering the following related questions: (a) How do the C-T ERP estimates compare with each other? (b) How do the performances of nearest-estimate classifiers compare with each other? (c) For a given ERP channel, do the heterogeneous nearest-estimate classifiers offer complementary information for improving performance through decision fusion? (d) Do the homogeneous nearest-estimate classifiers of different channels offer complementary information for improving performance through decision fusion? (e) Can the performance be improved by fusing the decisions of all or a selected subset of the entire classifier ensemble? These questions are answered by designing estimation and classification experiments using real 6-channel ERPs. It is shown that although the operations to compute the vector C-T estimates can be quite different, the ERP estimates are similar with respect to their overall waveform shapes and peak latencies. Furthermore, the results of the classification experiments show that by fusing homogeneous nearest-estimate classifier decisions across multiple channels, the classification accuracy can be improved significantly when compared with the accuracies of individual channel classifiers.     

Boosting web video categorization with contextual information from social web

Web video categorization is a fundamental task for web video search. In this paper, we explore web video categorization from a new perspective, by integrating the model-based and data-driven approaches to boost the performance. The boosting comes from two aspects: one is the performance improvement for text classifiers through query expansion from related videos and user videos. The model-based classifiers are built based on the text features extracted from title and tags. Related videos and user videos act as external resources for compensating the shortcoming of the limited and noisy text features. Query expansion is adopted to reinforce the classification performance of text features through related videos and user videos. The other improvement is derived from the integration of model-based classification and data-driven majority voting from related videos and user videos. From the data-driven viewpoint, related videos and user videos are treated as sources for majority voting from the perspective of video relevance and user interest, respectively. Semantic meaning from text, video relevance from related videos, and user interest induced from user videos, are combined to robustly determine the video category. Their combination from semantics, relevance and interest further improves the performance of web video categorization. Experiments on YouTube videos demonstrate the significant improvement of the proposed approach compared to the traditional text based classifiers.     

Combining classifiers using nearest decision prototypes

We present a new classifier fusion method to combine soft-level classifiers with a new approach, which can be considered as a generalized decision templates method. Previous combining methods based on decision templates employ a single prototype for each class, but this global point of view mostly fails to properly represent the decision space. This drawback extremely affects the classification rate in such cases: insufficient number of training samples, island-shaped decision space distribution, and classes with highly overlapped decision spaces. To better represent the decision space, we utilize a prototype selection method to obtain a set of local decision prototypes for each class. Afterward, to determine the class of a test pattern, its decision profile is computed and then compared to all decision prototypes. In other words, for each class, the larger the numbers of decision prototypes near to the decision profile of a given pattern, the higher the chance for that class. The efficiency of our proposed method is evaluated over some well-known classification datasets suggesting superiority of our method in comparison with other proposed techniques.