基于主题分布优化的模糊文本分类研究

在对类别模糊的文本进行分类时,主题模型只考虑文档和主题级别信息,未考虑底层词语间的隐含信息,且多数主题信息复杂、中心不明确。为此,提出一种改进的文本分类方法。通过分位数选择中心明确的主题,将其映射到word2vec词向量空间内,对模糊文本进行分类操作,进而得到文本分类结果。实验结果表明,与C_LCD+KNN方法相比,该方法分类效果较好,鲁棒性较强。     

变精度粗集模型在决策树生成过程中的应用

Pawlak粗集模型所描述的分类是完全精确的,而没有某种程度上的近似。在利用Pawlak粗集模型构造决策树的过程中,生成方法会将少数特殊实例特化出来,使生成的决策树过于庞大,从而降低了决策树对未来数据的预测和分类能力。利用变精度粗集模型,对基于Pawlak粗集模型的决策树生成方法进行改进,提出变精度明确区的概念,允许在构造决策树的过程中划入明确区的实例类别存在一定的不一致性,可简化生成的决策树,提高决策树的泛化能力。     

基于人工神经网络与决策树相结合模型的遥感图像自动分类研究

本文提出了一种新的基于Kohonen神经网络与决策树相结合模型的遥感图像自动分类方法。选取绍兴地区为实验区，对TM图像进行了分类实验。并将该模型分类结果与基于Kohonen网络模型的分类结果进行了比较，发现对于江南低山丘陵河网密集区的TM图像应用该模型进行分类能够得到较为满意的分类结果，其分类精度可达到85．16％，较之单纯使用Kohonen网络模型提高了20．12％。     

结合DCTM与HMM的音乐分类方法

改进相关主题模型(correlated topic model,CTM)使其具有动态性,提出了动态相关主题模型(dynamic correla-ted topic model,DCTM),使用变分卡尔曼滤波推断模型的隐含主题参数。将DCTM作为降维模型与隐马尔克夫模型(hidden Markov model,HMM)相结合对音乐分类。这一方法将音乐片段分割为等长的小片段,将小片段的声学特征向量通过相似性比较转化为单词序列,通过DCTM将单词序列转换为主题向量。将主题向量输入HMM得出分类结果。由于DCTM的动态建模,更好地提取对分类有用的信息,因此增强了方法的分类能力。实验验证了方法的有效性。     

基于多核学习支持向量机的音乐流派分类

针对不同特征向量下选择最优核函数的学习方法问题,将多核学习支持向量机(MK-SVM)应用于音乐流派自动分类中,提出了将最优核函数进行加权组合构成合成核函数进行流派分类的方法。多核分类学习能够针对不同的声学特征采用不同的最优核函数,并通过学习得到各个核函数在分类中的权重,从而明确各声学特征在流派分类中的权重,为音乐流派分类中特征向量的分析和选择提供了一个清晰、明确的结果。在ISMIR 2011竞赛数据集上验证了提出的基于多核学习支持向量机(MKL-SVM)的分类方法,并与传统的基于单核支持向量机的方法进行了比较分析。实验结果表明基于MKL-SVM的音乐流派自动分类准确率比传统单核支持向量机的分类准确率提高了6.58%,且该方法与传统的特征选择结果比较,更清楚地解释了所选择的特征向量对流派分类的影响大小,通过选择影响较大的特征组合进行分类,分类结果也有了明显的提升。     

结合邻居辅助策略的两阶段层次文本分类模型

传统两阶段层次文本分类模型（THTC模型）是一种解决大规模层次文本分类问题的有效方法,但该模型的分类准确率仍然不是很高。为了缓解这个问题,提出了结合邻居辅助策略的两阶段层次文本分类模型（THTC-NA模型）。THTC-NA模型由搜索阶段和分类阶段组成。搜索阶段采用扁平策略从所有的叶子类别中选择与待分类文档最相关的[k]个类别作为候选类别集,这样可以大大减小分类阶段的搜索空间。分类阶段通过结合候选类别的祖先类别和兄弟类别的分类结果来帮助计算候选类别在分类阶段的结果。最后将搜索阶段的结果和分类阶段的结果融合起来共同决定待分类文档的目标类别。在数据集Newsgroups-18828上的实验表明,相对于THTC模型,THTC-NA模型对提高层次文本分类准确率有很大的帮助。     

基于多模态融合的三维模型检索算法研究

为了获得更好的三维模型检索分类性能,基于深度学习模型研究了多模态信息融合对三维模型的特征描述,在训练步骤提出相关性损失函数来指导不同模态之间的训练,提取更稳健的特征向量;最后将融合特征应用于三维模型的检索和分类,在ModelNet40数据集上进行了三维模型分类任务和检索任务评估。实验结果及与现有方法进行的对比证明了该方法的优越性,为三维模型检索分类领域提供了一种新的思路。     

基于情感模型的文本意见分类方法

基于向量空间模型、潜在语义分析等传统文本意见分类模型将文本映射到词汇或语义空间中,侧重于词汇的辨别能力,无法对映像空间给出明确的语义说明,导致其扩展性、准确率等方面的性能受到限制。为此,在人类情感分类理论的基础上,假设文本中的意见表达与人们的情感存在较强的关联,结合词汇语义扩展、特征选择等方法构造3种情感表示模型,把表达人类情感倾向的文本转换到情感空间中,利用情感模型对国外股票论坛信息提取情感特征,构建情感模型,并设计文本意见分类方法。针对实际股票论坛的数据进行实验,结果表明,该分类方法能获得较高的分类准确率。     

基于深度学习的民事案件判决结果分类方法研究

裁判文书数量的快速增长对自动化分类提出了迫切要求,然而已有研究缺乏在民事案件这一细分领域下以判决结果为分类标准的方法的研究,无法实现对民事案件判决结果的准确分类.文中将深度学习技术应用于民事案件判决结果分类领域,通过横向对比多种深度学习模型得出了该领域下表现较好的模型,并依据裁判文书的数据特点对该模型进行了进一步的优化.实验结果证明,Transformer模型的判决结果分类的宏精准率、宏召回率和宏F1分数均高于其他模型.通过对数据预处理流程的优化和对Transformer模型位置嵌入方式的优化,模型的性能指标提升了1％～2％.     

利用微调卷积神经网络的土地利用场景分类

针对场景类别之间的相同类内差异性与不同类间相似性所造成的遥感图像场景分类不够精确的问题,提出了将微调(fine-tuning)与卷积神经网络(convolutional neural network,CNN)模型相结合的方法,对土地利用遥感场景图像进行分类。该方法对CNN前层固定,调整分类层,保留了图像的泛性特征;通过卫星影像图获取土地利用场景图块作为训练样本,对训练样本图块进行预处理,然后对在ImageNet数据集上训练得到的AlexNet模型进行fine-tuning,利用得到的CNN模型即可自动提取土地利用遥感图像的图像特征并对其进行分类。为了验证本文方法,对实验区影像进行分割得到测试样本并进行同训练样本一致的预处理,将测试样本的分类结果与随机森林、支持向量机等经典方法的结果进行对比。结果表明,经过fine-tuning的CNN模型在土地利用分类中得到的结果要明显优于其他分类方法。     

A Self-Organizing TS-Type Fuzzy Network With Support Vector Learning and its Application to Classification Problems

A self-organizing Takagi-Sugeno (TS)-type fuzzy network with support vector learning (SOTFN-SV) is proposed in this paper. The proposed SOTFN-SV is inspired by analysis of TS-type fuzzy systems and composite-kernel support vector machine (SVM). SOTFN-SV is a fuzzy system constructed by the hybridization of fuzzy clustering and SVM. The antecedent part of SOTFN-SV is generated via fuzzy clustering of the input data, and then SVM is used to tune the consequent part parameters to give the network better generalization performance. For demonstration, SOTFN-SV is applied to several classification problems, especially the skin color classification problem. In the skin color classification application, each color pixel is represented by hue and saturation (HS) color space. To represent color information by histogram as accurately as possible, a nonuniform partition of HS space is proposed. For comparison, SVMs and other fuzzy systems trained by SVM or neural networks are applied to the same classification problems. The advantages of SOTFN-SV are verified by comparisons with the results of these methods.     

基于模糊高斯基函数神经网络的遥感图像分类

针对遥感图像分类的特点,提出了一种基于模糊高斯基函数神经网络的遥感图像分类器。该分类器将模糊技术与神经网络相结合,采用神经网络来实现模糊推理,利用神经网络的学习能力来达到调整模糊隶属函数和模型规则的目的,从而使系统具备了自适应的特性,实验结果表明,这种基于模糊高斯基孙数神经网络的分类器经过训练后,可应用于遥感图像的分类,其分类精度明显高于传统的最大似然分类法。     

基于特征属性分类改进的模糊聚类方法及其应用

模糊聚类分析是一种重要的分类方法。传统模糊聚类分析法着眼于全体属性,在对多属性数据集分类方面具有明显优势,对基于特定、重要属性的分类时显得不足。本文对传统方法进行改进,提出了一种基于特征属性分类的模糊聚类方法,利用特征属性进行分类,产生了较好的分类效果,展示了一个成用实例。改进的方法人人提高了特定分类问题的应用价值。     

Improving urban classification through fuzzy supervised classification and spectral mixture analysis

In this study, a fuzzy‐spectral mixture analysis (fuzzy‐SMA) model was developed to achieve land use/land cover fractions in urban areas with a moderate resolution remote sensing image. Differed from traditional fuzzy classification methods, in our fuzzy‐SMA model, two compulsory statistical measurements (i.e. fuzzy mean and fuzzy covariance) were derived from training samples through spectral mixture analysis (SMA), and then subsequently applied in the fuzzy supervised classification. Classification performances were evaluated between the ‘estimated’ landscape class fractions from our method and the ‘actual’ fractions generated from IKONOS data through manual interpretation with heads‐up digitizing option. Among all the sub‐pixel classification methods, fuzzy‐SMA performed the best with the smallest total_MAE (MAE, mean absolute error) (0.18) and the largest Kappa (77.33%). The classification results indicate that a combination of SMA and fuzzy logic theory is capable of identifying urban landscapes at sub‐pixel level.     

GA-fuzzy modeling and classification: complexity and performance

The use of genetic algorithms (GAs) and other evolutionary optimization methods to design fuzzy rules for systems modeling and data classification have received much attention in recent literature. Authors have focused on various aspects of these randomized techniques, and a whole scale of algorithms have been proposed. We comment on some recent work and describe a new and efficient two-step approach that leads to good results for function approximation, dynamic systems modeling and data classification problems. First, fuzzy clustering is applied to obtain a compact initial rule-based model. Then this model is optimized by a real-coded GA subjected to constraints that maintain the semantic properties of the rules. We consider four examples from the literature: a synthetic nonlinear dynamic systems model, the iris data classification problem, the wine data classification problem, and the dynamic modeling of a diesel engine turbocharger. The obtained results are compared to other recently proposed methods     

An efficient neural classification chain of SAR and optical urban images

In this paper a suitable neural classification algorithm, based on the use of Adaptive Resonance Theory (ART) networks, is applied to the fusion and classification of optical and SAR urban images. ART networks provide a flexible tool for classification, but are ruled by a large number of parameters. Therefore, the simplified ART2-A algorithm is used in this paper, and the neural approach is integrated into a classification chain where fuzzy clustering for merging of classes is also considered. The interaction between the two methods leads to encouraging results in less CPU time than classification with fuzzy clustering alone or other classical approaches (ISODATA). Examples of classification are provided using C-band total power AIRSAR data and optical images of Santa Monica, Los Angeles.     

Induction of descriptive fuzzy classifiers with the Logitboost algorithm

Recently, Adaboost has been compared to greedy backfitting of extended additive models in logistic regression problems, or “Logitboost". The Adaboost algorithm has been applied to learn fuzzy rules in classification problems, and other backfitting algorithms to learn fuzzy rules in modeling problems but, up to our knowledge, there are not previous works that extend the Logitboost algorithm to learn fuzzy rules in classification problems.In this work, Logitboost is applied to learn fuzzy rules in classification problems, and its results are compared with that of Adaboost and other fuzzy rule learning algorithms. Contradicting the expected results, it is shown that the basic extension of the backfitting algorithm to learn classification rules may produce worse results than Adaboost does. We suggest that this is caused by the stricter requirements that Logitboost demands to the weak learners, which are not fulfilled by fuzzy rules. Finally, it is proposed a prefitting based modification of the Logitboost algorithm that avoids this problem     

Compact fuzzy association rule-based classifier

Classification is one of the most popular data mining techniques applied to many scientific and industrial problems. The efficiency of a classification model is evaluated by two parameters, namely the accuracy and the interpretability of the model. While most of the existing methods claim their accurate superiority over others, their models are usually complex and hardly understandable for the users. In this paper, we propose a novel classification model that is based on easily interpretable fuzzy association rules and fulfils both efficiency criteria. Since the accuracy of a classification model can be largely affected by the partitioning of numerical attributes, this paper discusses several fuzzy and crisp partitioning techniques. The proposed classification method is compared to 15 previously published association rule-based classifiers by testing them on five benchmark data sets. The results show that the fuzzy association rule-based classifier presented in this paper, offers a compact, understandable and accurate classification model.     

基于模糊积分和遗传算法的分类器组合算法

将多个分类器进行组合能提高分类精度。基于模糊测度的Sugeno和Choquet积分具有理想的特性,因此该文利用其进行分类器组合。然而在实际中难以求得模糊测度。该文利用两种方法求取模糊测度,一是分类器对样本数据的分类能力,另一种是根据遗传算法。这两种方法均考虑了每个分类器对不同类的分类能力不同这一经验知识。实验中对UCI中的几个数据库进行了测试,同时将该组合方法应用于一多传感器融合工件识别系统。测试结果表明了该算法是一种计算简便、精度较高的分类器组合方法。     

一种基于支持向量机的模糊分类系统研究

提出一种基于支持向量机学习的模糊分类束纯模型．通过将支持向量机映射成等价的模糊分类系统,支持向量机的稀疏性表示等特性使得相应的模糊分类系统避免了“维数灾难”问题,并具有良好的泛化能力．另一方面,模糊系统的一些理论和应用成果也可用来进一步改善分类系统的性能．本文根据模糊集合的贴近度概念对模糊系统的语言变量进行约简,合并冗余的和不一致的模糊规则,然后采用粒子群优化方法改善模糊分类系统性能．该方法增强了系统的泛化能力,并可以理解为解决支持向量机中难以确定的系统参数问题的一种辅助方法．实验结果表明了该方法的可行性和有效性．