基于AUC的SVM多类分类方法的研究

AUC(ROC曲线下面积)评价标准已经广泛地用于度量机器学习中各种分类算法在两类数据集上的分类性能。首先介绍了SVM(支持向量机)多类分类方法,然后对AUC方法进行了系统地介绍,最后通过实验来比较各种SVM多类分类方法在多类别数据集上的AUC的值。实验结果表明,AUC值和核函数和多类转换方法的选取都有着密切的联系。

基于多类支持向量机递归特征消除方法特征选择的原发性肝癌患者预后预测

本研究通过特征选择的方法,分析肝癌患者术前临床信息,提高患者的预后模型的准确性。基于多类支持向量机递归特征消除(recursive feature elimination based on multiple support vector machine,MSVM-RFE)方法对进行过肝切除手术的原发性肝癌患者的临床变量进行重要特征排序,使用5折交叉验证的支持向量机确定最优特征子集,构造原发性肝癌患者术后的1年、3年无瘤生存和总体生存的列线图。通过与临床医生沟通,确认特征排序结果为合理的。患者3年无瘤生存风险和总生存风险的列线图的一致性指数分别为0.701和0.706。使用多类支持向量机递归特征消除方法后的预测模型准确率有所提高,列线图在临床实践中能够提供患者生存风险信息,简单清晰的反映患者的生存风险。

旋转机械故障智能诊断与预报方法综合研究

系统地介绍了由灰色理论、模糊理论、神经网络理论及模糊神经网络理论等各类先进的诊断技术汇总开发的一套旋转机械故障智能诊断与预报系统软件 ,实现了多类方法进行综合诊断的功能 ,并且通过实例验证了该诊断与预报系统软件的可行性 ,从而为旋转机械故障诊断与预报专家系统的研究与开发提供了一定的借鉴。

基于多示例学习框架的文本分类算法

针对有特殊结构的文本,传统的文本分类算法已经不能满足需求,为此提出一种基于多示例学习框架的文本分类算法。将每个文本当作一个示例包,文本中的标题和正文视为该包的两个示例;利用基于一类分类的多类分类支持向量机算法,将包映射到高维特征空间中;引入高斯核函数训练分类器,完成对无标记文本的分类预测。实验结果表明,该算法相较于传统的机器学习分类算法具有更高的分类精度,为具有特殊文本结构的文本挖掘领域研究提供了新的角度。

改进的SVM算法及其在故障诊断中的应用研究

介绍了支持向量机用于解决模式分类问题的基本原理,在对传统的多分类方法OVO(one-versus-one)深入分析的基础上,针对其存在的不可分类区问题,提出了一种改进的模式分类方法KSVM(KNN-SVM),将k-近邻方法嵌入到SVM算法中解决不可分类区问题,进一步提高了分类准确率。应用KSVM分类方法进行模拟电路的故障诊断,实验结果验证了该方法的有效性和实用性。

多种教学方式和自主学习相融合改进药学英语教学质量

药学英语作为中医药院校药学专业基础课程,是扩展学生学习范畴,提升学生学习层次的一门重要课程。本文结合药学专业学生的实际情况和药学英语教学目标,开展多种形式的教学和自主学习方法应用于药学专业英语教学中,激发学生学习兴趣,培养学生学习的积极性和主动性,旨在提高药学英语课程教学质量,让学生提升药学英语应用能力。

无线传感器网络中基于SVM的合作型入侵检测系统

由于许多重要的无线传感器网络需要一个高效、轻量级、灵活的入侵检测算法来检测恶意节点,提出了一种基于二叉树的SVM多类分类方法的合作型入侵检测方案。该方案可扩展性较好,有效地节省了传感器节点的能量。仿真实验表明,与已经提出的入侵检测系统相比,该方案具有较低的误报率和较高的检测率。

CLASSIFIER FUSION BASED ON EVIDENCE THEORY AND ITS APPLICATION IN FACE RECOGNITION

A multiple classifier fusion approach based on evidence combination is proposed in this paper. The individual classifier is designed based on a refined Nearest Feature Line (NFL),which is called Center-based Nearest Neighbor (CNN). CNN retains the advantages of NFL while it has relatively low computational cost. Different member classifiers are trained based on different feature spaces respectively. Corresponding mass functions can be generated based on proposed mass function determination approach. The classification decision can be made based on the combined evidence and better classification performance can be expected. Experimental results on face recognition provided verify that the new approach is rational and effective.

From Rubbish to a Large Scale Industry： A Simple Fabrication of Superfiber with Multitasking Applications

In the whole earth, people increased dramatically from generation to generation which had created a large scale of broken environment so that people are facing more various types of garbage. Most of garbages are not useful and as a matter of fact, they are used to be neglected. Furthermore, many efforts have been conducted to change it by many types of recycled methods. Here, a simple technique is proposed with and without using fires to transform the useless natural or man-made rubbish things to be a superfiber as well as thin film with multitasking applications in human daily life. Since most of earth environment is covered by oceans, here the authors show how the ocean related garbage such as the crab skins, broken coral reefs and beach stones were changed to be superfiber and a multitasking device prototype.

Multi-criteria Decision of Type and Building Material for Simple House Construction

Facing the innovation technology in building construction, the designer has the option in selecting the type of system construction and building materials. There is a big question for the designer and owner on how to evaluate the quality material to meet the technical requirements as building construction reliability. By demand to provide a low cost house, there is need for improving structure model and material construction for residents. When material cost takes 60%-70% of total construction cost, it is important to select the appropriate building material. Currently the innovation of material building for main frame and wall component has present for residential projects in Indonesia. Attempt to replace the traditional material such as brick or pozzolan lime concrete block is developed a new material include utilization of coal combustion waste for concrete block, lightweight concrete, or prefab panel wall. This paper presents the research result to develop a tool support for material and system construction selecting that facilitates the selection an optimal material for a simple house construction. A multi-criteria decision method is used based on performance criteria such as economy, reliability, comfort and eco-friend.

Limitations of the use of MTT assay for screening in drug discovery

The 3-（4,5-dimethylthiazol-2-yl）-2,5-diphenyl tetrazolium bromide （MTT） assay is a routine cell viability assay for cell proliferation and cytotoxicity, which is widely used in many fields, especially in screening for drug discovery. However, this assay exhibits limitations in the presence of particular compounds and under certain assay conditions, which may yield false screening results. For example, polyphenols that are extracted from natural sources can react with MTT in the absence of living cells and thus interfere with the screening results. We measured the absorbance of 15 polyphenols extracted from green tea and showed that the phenolic hydroxyl groups in the polyphenols are responsible for the reduction of MTT to formazan. When three or more phenolic hydroxyl groups were present on a conjugated polyphenol, a significantly increased MTT reduction was observed. Moreover, the type of medium also had an effect on the absorbance value, in the following order： ct-MEM ＋ 10% FBS〉 a-MEM〉DMEM/F12〉PBS. The absorbance of the MTT assay recorded at 570 nm is more sensitive than that measured at 595 nm. These results will improve the cell-based assay of polyphenols and clarify the limitations of the MTT assay as a method of screening in drug discovery.

High-resolution remote sensing mapping of global land water

Land water, one of the important components of land cover, is the indispensable and important basic information for climate change studies, ecological environment assessment, macro-control analysis, etc. This article describes the overall study on land water in the program of global land cover remote sensing mapping. Through collection and processing of Landsat TM/ETM+, China's HJ-1 satellite image, etc., the program achieves an effective overlay of global multi-spectral image of 30 m resolution for two base years, namely, 2000 and 2010, with the image rectification accuracy meeting the requirements of 1:200000 mapping and the error in registration of images for the two periods being controlled within 1 pixel. The indexes were designed and selected reasonably based on spectral features and geometric shapes of water on the scale of 30 m resolution, the water information was extracted in an elaborate way by combining a simple and easy operation through pixel-based classification method with a comprehensive utilization of various rules and knowledge through the object-oriented classification method, and finally the classification results were further optimized and improved by the human-computer interaction, thus realizing high-resolution remote sensing mapping of global water. The completed global land water data results, including Global Land 30-water 2000 and Global Land 30-water 2010, are the classification results featuring the highest resolution on a global scale, and the overall accuracy of self-assessment is 96%. These data are the important basic data for developing relevant studies, such as analyzing spatial distribution pattern of global land water, revealing regional difference, studying space-time fluctuation law, and diagnosing health of ecological environment.

An Improvement to Multiple Criteria ABC Inventory Classification Using Shannon Entropy

This paper extends the Ng-model [Ng, 2007] for multiple criteria ABC inventory classification based upon Shannon entropy. The proposed approach determines the common weights associated with all criteria importance rankings, and provides a comprehensive scoring scheme by aggregating all rankings of the criteria importance. A numerical illustration is presented to compare the model with previous studies.

Dynamically Optimized Sensor Deployment Based on Game Theory

Sensor network deployment is the key for sensors to play an important performance. Based on game theory, first, the authors propose a multi-type sensor target allocation method for the autonomous deployment of sensors, considering exploration cost, target detection value, exploration ability and other factors. Then, aiming at the unfavorable environment, e.g., obstacles and enemy interference, the authors design a method to maintain the connectivity of sensor network, under the conditions of effective detection of the targets. Simulation result shows that the proposed deployment strategy can achieve the dynamic optimization deployment under complex conditions.

Risk Assessment of Metal Contamination in Soil and Groundwater in Asia： A Review of Recent Trends as well as Existing Environmental Laws and Regulations

Asia is experiencing a more rapid economic growth compared to any other regions. The contamination of soil and groundwater with metals can mainly be attributed to human activities; therefore, risk assessments to characterize the nature and magnitude of risks to humans and ecological receptors from contaminants are important. Risk assessments are often iterative processes, which involve identification and filling data gap. Experimental samplings, geostatistical and multivariate statistical methods as well as multimedia risk assessment modeling are the three major methodologies used in the assessment of metal contamination in soil and groundwater.This review highlights a number of measurements for improving risk calculation methods and expounds scientific approaches that involve the identification of the major source of contamination, exposure pathways and bioavailability of metals. In general, risk assessments of metals in soil and groundwater worldwide are mainly focused on the levels of contamination, identification of exposure pathways, and prediction of the probability of contamination. To date, very limited studies have reported the development of relevant environmental laws and policies in the regulation of soil and groundwater contamination in Asia. The development, variations and limitations in the regulations of soil and groundwater contamination among developed countries may provide helpful guidance for the developing countries in Asia.

MsFEM à la Crouzeix-Raviart for Highly Oscillatory Elliptic Problems

We introduce and analyze a multiscale finite element type method (MsFEM) in the vein of the classical Crouzeix-Raviart finite element method that is specifically adapted for highly oscillatory elliptic problems. We illustrate numerically the efficiency of the approach and compare it with several variants of MsFEM.