基于实例的基因分类及确定基因标签模型

所建立的模型及所得的结论有利于利用数据库中已有的基因信息快速筛选出潜在的癌症相关基因，模型一和模型二以基因表达水平限值和差异显著性水平为分类要素，将基因分为两类．模型三利用逐步优化思想建立优化模型，确定出六组基因标签．模型四利用小波分析法去噪及相关性检验法，重新确定基因标签，包含8种特征基因，对癌症样本的检测率降低了，...     

基于高斯混合和BP神经网络的卵巢癌质谱数据三分类模型

癌症的早期诊断可以显著提高癌症患者的存活率,三分类问题就是将未知样本与已知样本进行匹配度检测,预测样本是健康状态,良性发展状态,还是癌症状态.针对复杂难分的卵巢癌蛋白质质谱数据,提出了一种基于高斯混合模型和BP神经网络的三分类预测模型.首先,去除原数据中的冗余,对其进行方差排序及交集筛选提取特征集合一,再利用高斯混合模型处理求得参数作为特征集合二,最后使用BP神经网络进行样本三分类,准确率达到72.9%.结果表明:模型可以作为卵巢癌质谱数据三分类的可选择工具.     

基于基因表达谱的癌症亚型分类模型研究

随着机器学习和生物信息学的快速发展,癌症亚型分类成为当前研究热点之一.根据亚型的分类,可以指导癌症的治疗和预后.近年来,许多监督学习方法被用于癌症亚型分类.考虑到高维、样本数量少和数据不均衡等特点,本文首先利用LDA进行降维,其次利用SMOTE算法均衡数据,再利用Extra-Trees模型对癌症亚型进行分类,最后基于TCGA中9种癌症25种癌症亚型的3 296个样本来验证模型的有效性.实验结果表明,利用给出的模型进行癌症亚型分类具有很好的效果.     

约束含参分岔问题的分类

约束边界与分岔参数有关的约束分岔问题,称为约束含参分岔问题．通过引入适当的变换,将约束含参分岔问题转化为新变量的非约束分岔问题,推导出了约束含参分岔问题转迁集的一般形式,结果表明只有约束分岔集受约束含参的影响,其它转迁集与不含参约束分岔的转迁集相同．以含参约束树枝分岔为例分析了此类问题的分岔分类,讨论了约束含参对分岔分类的影响．     

基于贝叶斯统计方法的两总体基因表达数据分类

在疾病的诊断过程中,对疾病的精确分类是提高诊断准确率和疾病治愈率至 关重要的一个环节,DNA芯片技术的出现使得我们从微观的层次获得与疾病分类及诊断 密切相关的基因功能信息．但是DNA芯片技术得到的基因的表达模式数据具有多变量小 样本特点,使得分类过程极不稳定,因此我们首先筛选出表达模式发生显著性变化的基因 作为特征基因集合以减少变量个数,然后再根据此特征基因集合建立分类器对样本进行分 类．本文运用似然比检验筛选出特征基因,然后基于贝叶斯方法建立了统计分类模型,并 应用马尔科夫链蒙特卡罗(MCMC)抽样方法计算样本归类后验概率．最后我们将此模型 应用到两组真实的DNA芯片数据上,并将样本成功分类．     

基于满意度函数的多响应稳健优化模型及实证研究

针对传统的综合满意度模型在研究多响应时较少考虑噪声因素的问题,本文通过引入噪声因素对传统的满意度模型实施改进,进行多响应稳健优化分析。首先,构建包含可控因素和噪声因素的多质量特性的响应曲面模型;其次,借助传统综合满意度模型的构建方法,将多质量特性的响应曲面模型整合为改进的满意度模型;最后,将信噪比作为衡量改进的满意度模型稳健优化的指标,得到稳健优化的参数组合。实证研究表明:构建同时包含可控因素和噪声因素的改进的满意度模型是可行的,在此模型的基础上利用信噪比能够有效地找到多响应稳健优化的参数组合。     

多基站下基于信号到达时间的室内三维定位

对基于无线电信号的到达时间(TOA)的定位技术进行了研究.首先,建立了TOA测量值与真实值间的函数关系,通过求解含惩罚项的非线性最小二乘问题实现终端定位和噪声估计,并在已有研究的基础上提出了Alternating Iterative Reweighting(AIR)算法;其次,利用基站之间的几何位置进行筛选,根据每个终端信息完成自适应的最少基站的选择及定位;并结合物理特性实现终端运动轨迹预测;最后,根据基站通信半径筛选出可被定位终端,完成定位精度与终端平均连接度数之间关系的建模.     

基于神经网络MIV值分析的肿瘤基因信息提取

运用统计学及数据挖掘相关知识,以结肠癌基因表达图谱为研究对象,综合运用GB指数、BP神经网络、小波变换等方法对问题给出求解的过程和结果.首先采用GB综合指数对无关基因进行筛选,选择两组备用基因的交集(114个)作为信息基因,降低基因维度.其次,用基因间的强相关性剔除冗余基因,利用BP神经网络对基因进行错判数计算,选取错判率最低、基因子集中基因数量最少的基因特征组,再利用平均影响值(MIV)方法进行基因筛选,最后进行错判数计算,最终确定含有12个基因的子集为最优基因组合.第三步,将每组基因表达值看做基因信号,运用小波转换法对基因数据进行去噪,去噪后特征基因减少为8个.     

基于基因表达子宫内膜癌的分类

建立了四类基于基因表达的分类器,用以将87名妇女的子宫内膜样本分成癌症患者和非癌症患者.首先利用信噪比过滤掉无关基因,然后利用主成分分析降低样本维数,再针对这四类分类器随机取75个样本作为训练样本,其余的12个样本作为测试样本,实验结果表明这四类分类器适合子宫内膜癌的分类.最后采用留一交叉验证作为评判标准,通过比较,说明5BP-ELMAN分类器是一类更适合子宫内膜癌分类的有效的肿瘤分类器.     

机器学习方法在储层分类中的应用

油气田开发中有效储层和非有效储层的样本点存在混合带时,两类储层的划分是一个难点问题.从统计学上来看,其本质是一个含噪声的小样本二分类问题,可以采用机器学习方法,充分挖掘有试油成果的样本点的数据信息.分别利用线性判别分析、支持向量机、多层感知机神经网络建立储层分类模型,利用10次10折交叉验证法进行模型评估与优选,并利用全部样本点建立了有效的储层分类模型,最后将模型推广应用到样本分布的三种不同情形.结果表明,线性支持向量机模型具有最好的分类效果和很强的泛化能力,对于区分有效储层和非有效储层是有效的,可以在油气田开发中进行推广.     

基因分类的统计方法

基因组学进行肿瘤分型中的基因筛选方法进行探讨,根据相关分析和95%参考值范围的概念确定筛选基因的方法,该方法利用了所有基因的信息,计算简便,在对白血病患者分型时取得了很好的效果.     

Bayesian variable selection with sparse and correlation priors for high-dimensional data analysis

The main challenge in working with gene expression microarrays is that the sample size is small compared to the large number of variables (genes). In many studies, the main focus is on finding a small subset of the genes, which are the most important ones for differentiating between different types of cancer, for simpler and cheaper diagnostic arrays. In this paper, a sparse Bayesian variable selection method in probit model is proposed for gene selection and classification. We assign a sparse prior for regression parameters and perform variable selection by indexing the covariates of the model with a binary vector. The correlation prior for the binary vector assigned in this paper is able to distinguish models with the same size. The performance of the proposed method is demonstrated with one simulated data and two well known real data sets, and the results show that our method is comparable with other existing methods in variable selection and classification.     

Benchmarking state-of-the-art classification algorithms for credit scoring

In this paper, we study the performance of various state-of-the-art classification algorithms applied to eight real-life credit scoring data sets. Some of the data sets originate from major Benelux and UK financial institutions. Different types of classifiers are evaluated and compared. Besides the well-known classification algorithms (eg logistic regression, discriminant analysis, k-nearest neighbour, neural networks and decision trees), this study also investigates the suitability and performance of some recently proposed, advanced kernel-based classification algorithms such as support vector machines and least-squares support vector machines (LS-SVMs). The performance is assessed using the classification accuracy and the area under the receiver operating characteristic curve. Statistically significant performance differences are identified using the appropriate test statistics. It is found that both the LS-SVM and neural network classifiers yield a very good performance, but also simple classifiers such as logistic regression and linear discriminant analysis perform very well for credit scoring.     

函数S-粗集与规律双向分离

函数S-粗集(function singular rough sets)是用R-函数等价类定义的,函数是一个规律,函数S-粗集具有规律特征.函数S-粗集推广了Z.Pawlak粗集.利用函数S-粗集,给出规律生成,规律分离的讨论,提出规律分离定理.给出的结果在投资分险规律估计中得到了应用.     

On the Euler characteristic of finite unions of convex sets

The Euler characteristic plays an important role in many subjects of discrete and continuous mathematics. For noncompact spaces, its homological definition, being a homotopy invariant, seems not as important as its role for compact spaces. However, its combinatorial definition, as a finitely additive measure, proves to be more applicable in the study of singular spaces such as semialgebraic sets, finitely subanalytic sets, etc. We introduce an interesting integral by means of which the combinatorial Euler characteristic can be defined without the necessity of decomposition and extension as in the traditional treatment for polyhedra and finite unions of compact convex sets. Since finite unions of closed convex sets cannot be obtained by cutting convex sets as in the polyhedral case, a separate treatment of the Euler characteristic for functions generated by indicator functions of closed convex sets and relatively open convex sets is necessary, and this forms the content of this paper.     

Employment of neural network and rough set in meta-learning

The selection of the optimal ensembles of classifiers in multiple-classifier selection technique is un-decidable in many cases and it is potentially subjected to a trial-and-error search. This paper introduces a quantitative meta-learning approach based on neural network and rough set theory in the selection of the best predictive model. This approach depends directly on the characteristic, meta-features of the input data sets. The employed meta-features are the degree of discreteness and the distribution of the features in the input data set, the fuzziness of these features related to the target class labels and finally the correlation and covariance between the different features. The experimental work that consider these criteria are applied on twenty nine data sets using different classification techniques including support vector machine, decision tables and Bayesian believe model. The measures of these criteria and the best result classification technique are used to build a meta data set. The role of the neural network is to perform a black-box prediction of the optimal, best fitting, classification technique. The role of the rough set theory is the generation of the decision rules that controls this prediction approach. Finally, formal concept analysis is applied for the visualization of the generated rules.     

Estimation of fractal dimension and fractal curvatures from digital images

Most of the known methods for estimating the fractal dimension of fractal sets are based on the evaluation of a single geometric characteristic, e.g. the volume of its parallel sets. We propose a method involving the evaluation of several geometric characteristics, namely all the intrinsic volumes (i.e. volume, surface area, Euler characteristic, etc.) of the parallel sets of a fractal. Motivated by recent results on their limiting behavior, we use these functionals to estimate the fractal dimension of sets from digital images. Simultaneously, we also obtain estimates of the fractal curvatures of these sets, some fractal counterpart of intrinsic volumes, allowing a finer classification of fractal sets than by means of fractal dimension only. We show the consistency of our estimators and test them on some digital images of self-similar sets.     

Function S-rough Sets and Their Law Characteristic

Function S-rough sets(Function Singular rough sets) are defined by R-function equivalence class which has dynamic characteristic, and a function is s law, function S-rough sets have law characteristic. Function S-rough sets has these forms: function one direction S-rough sets, function two direction S-rough sets and dual of function one direction S-rough sets. This paper presents the law characteristic of function one direction S-rough sets and puts forward the theorems of law-chain-attribute and law-belt. Function S-rough sets is s new research direction of the rough sets theory.     

Generic bifurcations of varieties II

In the previous paper [3], we have developed a smooth classification theory of zero point sets of parametrized smooth map germs. In this paper we study a topological classification theory. It is closely related to Y. H. Wan's theory in [5].