共查询到20条相似文献，搜索用时 125 毫秒
1.
为提高分类性能,提出了一种新的基于数据离散化和选择性集成的SVM集成学习算法。该算法采用粗糙集和布尔推理离散化方法处理数据集,构造有差异的个体SVM以提高集成学习的性能。在训练得到一批SVM之后,算法采用了选择性集成提高性能并减小集成规模。实验结果表明,所提算法能取得比传统集成学习方法Bagging和Adaboost更好的性能。 相似文献
2.
3.
Parameters selection in gene selection using Gaussian kernel support vector machines by genetic algorithm 总被引：2，自引：0，他引：2
INTRODUCTION Recent techniques based on oligonucleotide or cDNA microarrays allow the expression level of thousands of genes to be monitored in parallel (Golub et al., 1999). A critically important factor for cancer diagnosis and treatment is the reliable prediction of tumor progression. A remarkable advance for mo- lecular biology and for cancer research is cDNA mi- croarray technology. cDNA microarray datasets havea high dimensionality corresponding to the large number of genes monit… 相似文献
4.
黄国华 《湖南第一师范学报》2011,11(2):133-136
以支持向量机为分类器,序列的k-letter词为特征,建立了原核生物的基因识别模型。分别选取已知功能的基因为正样本,和与等长正样本的随机突变序列为负样本组成训练集。5倍交叉实验的结果表示,对于具有不同核函数的支持向量机以及不同长度的词特征,其预测准确率不同,最高的可达94％以上,最差的低于60％;长度为3的词的特征的分类结果最好,其次是长度为4。这说明3联核苷酸为基因序列比较好的统计特征。 相似文献
5.
RAHIMI N SADEGHI M. H. MAHJOOB M.J. 《浙江大学学报(A卷英文版)》2007,8(9):1443-1451
In this paper,a geometric approach to fault detection and isolation (FDI) is applied to a Multiple-Input Multi-ple-Output (MIMO) model of a frame and the FDI results are compared to the ones obtained in the Single-Input Single-Output (SISO),Multiple-Input Single-Output (MISO),and Single-Input Multiple-Output (SIMO) cases. A proper distance function based on parameters obtained from parametric system identification method is used in the geometric approach. ARX (Auto Regressive with eXogenous input) and VARX (Vector ARX) models with 12 parameters are used in all of the above-mentioned models. The obtained results reveal that by increasing the number of inputs,the classification errors reduce,even in the case of applying only one of the inputs in the computations. Furthermore,increasing the number of measured outputs in the FDI scheme results in decreasing classification errors. Also,it is shown that by using probabilistic space in the distance function,fault diagnosis scheme has better performance in comparison with the deterministic one. 相似文献
6.
针对当前基于LDA模型扩展的文本情感分析方法未能考虑同一词语在不同语境下其情感极性的差异及非特征情感词对微博文本情感极性的影响这两个问题,提出一种基于语境分类和遗传算法的微博情感分析方法。该方法首先利用LDA模型构造微博主题集及微博主题词集,然后用微博标签数据逐一对各微博主题词集应用遗传算法自动迭代计算得出词集中词语的情感值,最后利用词集词语的情感值计算微博文本情感极性。实验结果表明,该方法精确度比LDA提升3.12%,召回率达87.32%,F1达73.79%,能够从语境和非特征情感词获取微博情感信息,有效提高情感分类准确率。 相似文献
7.
评估指标体系的选取是企业信用评估的首要问题，它是一个特征选择问题。文章提出了一种针时SVM组合技术的拆分特征选择方法，其主要思想是时SVM组合中的各个分类器分别进行特征选择，再采用不同的特征子集作为各子分类器的输入，进行组合建模与预测。文章从filter和wrapper相结合的思想出发，进行了子分类器的特征选择；之后，针对企业信用评估问题的特点，采用了二叉树结构作为SVM的组合策略。实验表明，拆分特征选择方法能选出规模较小、具有一定差异的关键指标集，提高了模型的分类性能，并且具有计算简单，运行快速的优点。 相似文献
8.
This article reports the results of a study examining the predictive validity of a computer attitude instrument. The researchers attempted to determine the extent to which this instrument predicts student learning. Data from two universities were collected using this instrument over a nine-year period and were sorted into three sets with a random n of 400 in each. Three procedures were performed. First, one set of data was used to develop a base model of prediction. Second, this model was used to calculate the predicted learning achievement scores for the other two sets of data. Finally, in those two sets of data, the means of the predicted and observed learning achievement scores were compared using inferential statistics. The predictive validity of the instrument was confirmed, as no significant differences were found between the mean predicted and observed learning outcome scores. 相似文献
9.
Robert G. MacCann Gordon Stanley 《Assessment in Education: Principles, Policy & Practice》2010,17(3):255-272
In educational systems, concern has been expressed about the accuracy of classification when marks are aligned to grades or levels. In particular, it has been claimed that a school assessment‐based grading would have much greater levels of accuracy than one based on examination scores. This paper investigates classification consistency by analysing five years of examination and assessment data in the subject areas of English and mathematics, and creating simulated parallel‐test observed scores at varying reliabilities (based on classical test theory assumptions). While grades created from moderated school assessments did show greater agreement than those from examination scores, the improvement was only modest. 相似文献
10.
为了提升不平衡数据中少数类的分类精度,利用SMOTE采样方法对数据集进行平衡化预处理;为了减轻样本重新合成过程中产生的类重叠和噪声对分类精度的影响,选择模糊粗糙最近邻算法（FRNN）作为分类器。在14个不平衡数据集上进行的仿真实验表明,该方法具有较好的分类表现,F值和G值最高分别可达0.965、0.932,是一种适用于不平衡率偏高数据集的分类方法。 相似文献
11.
基于粗糙集理论的电力电子电路故障诊断 总被引：1，自引：0，他引：1
基于粗糙集理论，提出了一种电力电子电路故障诊断的新方法：粗糙集方法。简单介绍了粗糙集理论，详细叙述了利用粗糙集理论诊断电力电子电路的过程。文章以三相可控整流电路为例，利用粗糙集理论，对故障信息中样本的故障征兆进行处理，通过知识约简，删除多余的征兆属性，形成诊断的确定性规则，实现故障分类。仿真实例表明，与以往的诊断方法比较，该方法不仅准确可靠，而且直观方便。 相似文献
12.
Fabio Alivernini 《Educational studies》2013,39(4):399-417
The aim of the present study, based on data from 20 countries, is to identify the pattern of variables (at country, school and student levels), which are typical of students performing below the low international benchmark compared to students performing at the advanced performance benchmark, in the Progress in International Reading Literacy Study (PIRLS) 2006. The dependent variable of the analysis is a dichotomous variable, the values of which represent the two different performance groups of students. The independent variables are two sets of Organisation for Economic Co-operation and Development educational indicators, variables from PIRLS 2006 Reading Curriculum Questionnaire data, variables and indices based on data obtained from questionnaires for teachers, schools, parents and students. The analysis is based on classification and regression trees, which is a full hierarchical non-parametric method suited to detecting and interpreting complex reciprocal influences between a large number of independent variables. The results show that pupils’ performance can be predicted by the relationships between country-level variables (the changes in teachers’ salaries at the top of the salary scale), school context variables (e.g. percentages of students from economically affluent home) and home educational resources and variables at student level (e.g. students’ reading self-concept). 相似文献
13.
《Structural equation modeling》2013,20(2):213-232
In the past, several models have been developed for the estimation of the reliability and validity of measurement instruments from multitrait-multimethod (MTMM) experiments. Suggestions have been made for additive, multiplicative and correlated uniqueness models, whereas recently Coenders and Saris (2000) suggested a procedure to test these models against one another. In this article, the different models suggested for the analysis of MTMM matrixes have been compared for their fit to 87 data sets collected in the United States (Andrews, 1984; Rodgers, Andrews, & Herzog, 1992), Austria (Koltringer, 1995), and the Netherlands (Scherpenzeel & Saris, 1997). As most variables are categorical, the analysis has been carried out on the basis of polychoric-polyserial correlation coefficients and of Pearson correlations. The fit of the models based on polychoric correlations is much worse than the fit of models based on product moment correlations, but in both cases a model that assumes additive method effects fits most data sets better than the other models, including the so-called multiplicative models. 相似文献
14.
把粗糙集与神经网络结合,应用于文本分类,可以充分发挥两种方法的优势,取长补短,粗糙集理论可以有效地对样本集进行约简,从而简化了神经网络的结构,减少了网络的训练次数,学习速度和分类精度明显提高,并用仿真实验验证了此方法的有效性. 相似文献
15.
陈永乐 《福建工程学院学报》2019,(6):606-612
构件族资源不足是建筑设计教学应用BIM技术具有代表性的教学问题之一。基于设计教学兼具正向设计和教学展示需求的特点,构件族开发遵循类型完整性、使用便捷性、模型重用率和过程集成性的思路,制定基于LOD300的分类、命名和模型细度标准,在对现有族库资源梳理整合的基础上,重点补充缺漏的构件类型,形成涵盖常用建筑构件及其基础信息的基础构件库。以幕墙族为例详述了基于参数化思维的建模方法,先按几何信息分布规律创建若干具有代表性的几何原型,再将消防、结构、节能设计等应用所需的非几何信息按类型与几何模型建立联动关系,形成各类可参变的幕墙构件族。 相似文献
16.
基于模式识别理论,将模糊集贴近度的概念引入大跨桥梁状态评估中,利用模糊层次分析,建立起大跨悬索桥多目标评估的数学模型.以润扬大桥悬索桥的综合状态评估为例,介绍了评估的基本程序,包括评估指标的层次划分、指标权重的确定、模糊集隶属函数的定义以及贴近度的计算等.评估综合了健康监测系统的采集数据和人工监测结果2方面的信息,同时还给出了评估指标的分类方法和确定性指标、不确定性指标的计算方法.与传统的专家打分法相比,该方法更好地体现了监测信息的离散性、不确定性和评估过程的模糊性. 相似文献
17.
Becker Mimi Larsen Congalton Russell G. Budd Rebecca Fried Alan 《Journal of Science Education and Technology》1998,7(1):85-96
Global Learning and Observations to Benefit the Environment (GLOBE) is an international environmental education and science partnership which coordinates the work of students (aged 5 to 18), teachers and scientists from 48 countries on five continents to study and better understand the global environment. Accurate ground reference data is fundamental to the use of remotely sensed data for land cover classification and mapping. Because very little ground reference data has been collected, the accuracy of many land cover maps may be questioned, thus accurate land cover ground reference data is an important need that could be addressed through GLOBE scientist-student collaboration. If earth systems scientists are to use student data, it is important that those data be as accurate as possible to ensure reliability of research results. Thus a key question for this research is whether student collected data are accurate enough to support rigorous scientific investigations. This paper describes results of the GLOBE Science-Education Team on Data Validation and Accuracy Assessment's collaboration with teachers and students to: (1) design and test the pre-protocol learning activities; (2) test the protocols intended to guide the collection and analysis of data; and (3) implement the learning activities and protocols to determine the relative accuracy of student collected versus professionally collected land cover data. To ensure the most accurate classification of land cover possible, a new international hierarchical land cover classification system, the Modified Unesco Classification (MUC) system was developed. GLOBE Data Collection Protocols and methods were designed and implemented to test the accuracy of student collected reference data were designed and implemented. Students who collected land cover reference data using GLOBE protocols, obtained data which are at least as accurate as that collected by professionals. 相似文献
18.
为提高大数据集粗分类识别率,提出一种基于聚类分析的SVM-Kd-tree树型粗分类方法。首先根据数据集特征分布进行k-means两簇聚类,对聚类后的数据集进行类别分析,同时将属于两簇的同一类别样本划分出来;然后使用两簇中剩余样本训练SVM二分类器并作为树型结构根节点,将两簇数据分别合并,将划分出来的样本作为左右子孩子迭代构建子节点,直到满足终止条件后,叶子节点开始训练Kd-tree。实验结果表明,迭代构建树型粗分类方法使训练单一SVM平均时间减少了61.977 4%,比Kd-tree同近邻数量的准确率提高了0.03%。在进行大规模数据集粗分类时,使用聚类分析迭代构建组合分类器时间更短、准确率更高。 相似文献
19.
遥感影像的统计分类中,通常都将像点特征的集合视为概率密度函数的混合分布,EM算法是求解这种混合模型参数的一个常用方法。但EM算法在给定合适初值的情况下,对训练数据中的噪声非常敏感,这将严重影响算法的运行效率和求取参数的精度。为了解决这个问题,本文提出了EM算法的模糊化策略,以此来减少噪声在参数学习过程中的影响。对遥感影像的分类实验表明,经过模糊化的EM算法能够更好地完成影像数据的分类。 相似文献
20.
Michael J. Beran James L. Pate W. Kirk Richardson Duane M. Rumbaugh 《Learning & behavior》2000,28(2):201-207
In this experiment, a chimpanzee’s (Pan troglodytes) long-term retention was examined. The chimpanzee, Lana, was trained to use lexigrams (geometric symbols representing linguistic units) for foods, colors, and objects when she was 2 years of age. At the age of 27, her recognition of three sets of lexigrams was examined. One of these sets (long term lexigrams) included five object lexigrams, one color lexigram, and one food lexigram, none of which had been seen by Lana for more than 20 years. The second set (different-referent lexigrams) contained lexigrams that are still present on current lexigram keyboards, but had been assigned new referents. The third set (same-referent lexigrams) contained lexigrams that had been kept on the keyboards Lana had used and that had retained the same referents. A food, a colored square, or an object was presented, and Lana had to select, by using a joystick, a lexigram on a computer screen. Lana chose the correct lexigram at a level significantly greater than chance for five of the seven lexigrams that she had not seen for more than 20 years. 相似文献