Exploring non-equilibrium statistical ensembles

Since Gibbs synthesized a general equilibrium statistical ensemble theory, generalizing the Gibbsian theory to non-equilibrium phenomena has been a dream for some theorists. However, the status of the theory for the non-equilibrium ensemble formalism is not as well established as the Gibbsian ensemble theory. In this work, we explore the formalism behind the non-equilibrium statistical ensemble theory based on the sub-dynamic kinetic equation originating from the Brussels-Austin school and develop...     

数据挖掘库预处理技术在教学中的应用研究

阐述了数据挖掘预处理技术的原理,结合教学实践研究了预处理的各种方法与技术实现,探讨了空缺值、不一致数据的处理,以及属性删除与构造,聚类技术进行归约的方法.     

旱涝灾害的遗传-神经网络集成预测方法研究

利用遗传算法的全局搜索能力同时进化设计三层BP神经网络的结构和连接权,并以进化后的网络结构和连接权作为新的神经网络结构和初始连接权,再进行新一轮附加动量的BP神经网络训练,把训练后的结果简单平均集成,以此建立旱涝灾害的遗传-神经网络集成预测新方法。应用该方法对广西桂林6月(主汛期1995~2005年)的降水量进行实例预测的结果表明,该方法的收敛速度快,预报精度高,易于操作,是一种具有较高应用价值的预测方法。     

基于二阶段异质随机森林的汽油辛烷值预测

构建了基于二阶段异质随机森林的汽油辛烷值预测模型.首先利用样本-位点信息表知识约简模型,筛选出对汽油辛烷值影响大的位点数据作为第一阶段;然后,利用集成学习思想集成支持向量回归和动态时间序列神经网络,构建异质随机森林预测模型作为第二阶段.利用十折交叉法验证模型精度,结果表明该集成学习算法具有有效性和高精度.     

Monte Carlo NPT系综法模拟考察几种水模型

应用NPT系综法模拟考察了五种水的势能模型 .模拟中采用周期边界条件、Metropolis抽样方法 ,并用Ewald方法对水分子间电荷作用进行长程校正 .模拟计算了内能 ,密度 ,压力及恒压热容等性质 .同时 ,还对这五种模型模拟得到的径向分布函数作了分析 .结果与文献及实验测定相比符合较好     

MS-IAS:集成的质谱代谢组学数据分析系统

针对代谢组学研究中的数据处理问题,本研究建立了基于质谱的数据分析系统MS-IAS(Mass spectrometry based integrated analysis system).此系统集成了特征选择、聚类、分类等多种方法,用以处理质谱数据,具有多种统计分析方法能对所选的特征变量进行比较,以发现与所研究问题相关的潜在生物标志物.MS-IAS支持数据与多种算法结果可图形化显示,有助于对数据的解释与分析.以肝病患者的质谱代谢组数据为例,展示MS-IAS的功能,两种特征选择算法从数据集中筛选出了40个对肝病具有区分能力的特征变量,展示了MS-IAS成为代谢组学研究中的通用质谱数据分析系统的潜力.     

气体混合物各组分化学势的Monte Carlo模拟

在则系综测试粒子Ｍｏｎｔｅ Ｃａｒｌｏ （ＧＣＭＣ）方法模拟常温下空气（以氮气为代表）及其污染物微量有机物（以苯为例）的混合物中各组分的化学势。模拟中,氮气和苯分子采用ＬＪ球型分子势能模型,采用Ｍｅｔｒｏｐｏｌｉｓ抽样及周期边界条件。通过模拟并拟合得到了３００．２Ｋ、苯的摩尔分数为０．０６２５,氮气及苯化学势与压力的关联式,以用于狭缝碳孔中该混合物体系的选择性吸附。     

Ensemble Approach for Detection of Depression Using EEG Features

Depression is a public health issue that severely affects one’s well being and can cause negative social and economic effects to society. To raise awareness of these problems, this research aims at determining whether the long-lasting effects of depression can be determined from electroencephalographic (EEG) signals. The article contains an accuracy comparison for SVM, LDA, NB, kNN, and D3 binary classifiers, which were trained using linear (relative band power, alpha power variability, spectral asymmetry index) and nonlinear (Higuchi fractal dimension, Lempel–Ziv complexity, detrended fluctuation analysis) EEG features. The age- and gender-matched dataset consisted of 10 healthy subjects and 10 subjects diagnosed with depression at some point in their lifetime. Most of the proposed feature selection and classifier combinations achieved accuracy in the range of 80% to 95%, and all the models were evaluated using a 10-fold cross-validation. The results showed that the motioned EEG features used in classifying ongoing depression also work for classifying the long-lasting effects of depression.     

Adaptive image enhancement based on clustering of wavelet coefficients for infrared sea surveillance systems

Most of the techniques developed for infrared (IR) image enhancement (IE) depend heavily on the scene, environmental conditions, and the properties of the imaging system. So, with a set of predefined scenario properties, a content-based IR-IE technique can be developed for better situational awareness. This study proposes an adaptive IR-IE technique based on clustering of wavelet coefficients of an image for sea surveillance systems. Discrete wavelet transform (DWT) of an image is computed and feature vectors are constructed from subband images. Clustering operation is applied to group similar feature vectors that belong to different scene components such as target or background. Depending on the feature vectors, a weight is assigned to each cluster and these weights are used to compute gain matrices which are used to multiply wavelet coefficients for the enhancement of the original image. Enhancement results are presented and a comparison of the performance of the proposed algorithm is given through subjective tests with other well known frequency and histogram based enhancement techniques. The proposed algorithm outperforms previous ones in the truthfulness, detail visibility of the target, artificiality, and total quality criteria, while providing an acceptable computational load.     

Pairwise clustering using a Monte Carlo Markov Chain

In this work an application of MCMC is proposed for unsupervised data classification, in conjunction with a novel pairwise objective function, which is shown to work well in situations where clusters to be identified have a strong overlap, and the centroid oriented methods (such as K-means) fail by construction. In particular, an exceptionally simple but difficult situation is addressed when cluster centroids coincide, and one can differentiate between the clusters only on the basis of their variance. Performance of the proposed approach is tested on synthetic and real datasets.