A new decision tree learning algorithm

Decision tree learning is one of the most widelyused and practical methods for inductive inference,andan effective approach to approximating discrete-valuedtarget functions,so decision tree learning is very suit-able for the multiclass classification.Up to now,deci-sion tree learning algorithms have been,by and large,classified as:1)univariant algorithms in which eachdecision node checks the value of a single attribute,such as ID3[1]and C4·5[2];and2)multivariant algo-rithms in which each deci…     

Optimization methods of cutting depth in mining Co-rich crusts

For optimizing the cutting depth of spiral drum type cutting head,the relations among collecting ratio,interfusing ratio of mullock and cutting depth of the mining cobalt-rich crusts in ocean were discussed.Furthermore,the multi-extremum problem about cutting depth was analyzed in mining at a certain interfusing ratio of mullock.Through introducing genetic algorithm(GA),the cutting depth-control problem when the collecting ratio is maximized by controlling the interfusing ratio of mullock was solved with global-optimization-search algorithms.Then optimization theory for cutting depth in mining cobalt-rich crusts by GA,and computer programming were given to realize the algorithm.The computation result of actual data proves the validity of this method.     

Feature Selection with Fluid Mechanics Inspired Particle Swarm Optimization for Microarray Data

Deoxyribonucleic acid (DNA) microarray gene expression data has been widely utilized in the field of functional genomics,since it is helpful to study cancer,cells,tissues,organisms etc.But the sample sizes are relatively small compared to the number of genes,so feature selection is very necessary to reduce complexity and increase the classification accuracy of samples.In this paper,a completely new improvement over particle swarm optimization (PSO) based on fluid mechanics is proposed for the feature selection.This new improvement simulates the spontaneous process of the air from high pressure to low pressure,therefore it allows for a search through all possible solution spaces and prevents particles from getting trapped in a local optimum.The experiment shows that,this new improved algorithm had an elaborate feature simplification which achieved a very precise and significant accuracy in the classification of 8 among the 11 datasets,and it is much better in comparison with other methods for feature selection.     

A complete discrimination system for polynomials with complex coefficients and its automatic generation

By establishing a complete discrimination system for polynomials, the problem of complete root classification for polynomials with complex coefficients is utterly solved, furthermore, the algorithm obtained is made into a general program in Maple, which enables the complete discrimination system and complete root classification of a polynomial to be automatically generated by computer, without any human intervention. Besides, by using the automatic generation of root classification, a method to determine the positive definiteness of a polynomial in one or two indeterminates is automatically presented.     

Online LS-SVM for function estimation and classification

An online algorithm for training LS-SVM (Least Square Support Vector Machines) was proposed for the application of function estimation and classification. Online LS-SVM means that LS-SVM can be trained in an incremental way, and can be pruned to get sparse approximation in a decremental way. When a SV (Support Vector) is added or removed, the online algorithm avoids computing large-scale matrix inverse. Thus the computation cost is reduced. Online algorithm is especially useful to realistic function estimation problem such as system identification. The experiments with benchmark function estimation problem and classification problem show the validity of this online algorithm.     

Segmentation and classification of high resolution imagery for mapping individual species in a closed canopy, deciduous forest

In this paper we investigate the use of a shadow-based delineation program for identifying segments in imagery of a closed canopy, deciduous forest, in West Virginia, USA, as a way to reduce the noise associated with per-pixel classification in forested environments. Shadows typically cluster along the boundaries of trees and therefore can be used to provide a network of nodes for the delineation of segments. A minimum cost path algorithm, where cost is defined as the cumulative sum of brightness values traversed along the connecting route, was used to connect shadow clumps. To test this approach, a series of classifications was undertaken using a multispectral digital aerial image of a six hectare test site and a minimum cost path segmentation. Three species were mapped: oaks, red maple and yellow poplar. The accuracy of an aspatial maximum likelihood classification (termed PERPIXEL classification) was 68.5%, compared to 74.0% for classification using the mean vector of the segments identified with the minimum cost path algorithm (MEAN_SEG), and 78% when the most common class present in the segment is assigned to the entire segment (POSTCLASS_SEG). By comparison, multispectral classification of the multispectral data using the field-mapped polygons of individual trees as segments, produced an accuracy of 82.3% when the mean vector of the polygon was used for classification (MEAN_TREE), and 85.7% when the most common class was assigned to the entire polygon (POSTCLASS_TREE). A moving window-based post-classification majority filter (POSTCLASS_MAJ5BY5) produced an intermediate accuracy value, 73.8%. The minimum cost path segmentation algorithm was found to correctly delineate approximately 28% of the trees. The remaining trees were either segmented, aggregated, or a combination of both segmented and aggregated. Varying the threshold that was used to discriminate shadows appeared to have little effect on the number of correctly delineated trees, or on the overall accuracy of the multispectral classification, although it did have a notable effect on the proportions of aggregated and segmented trees.     

Pattern Recognition and Forecast of Coal and Gas Outburst

Coal and gas outburst is a complicated dynamic phenomenon in coal mines, Multi-factor Pattern Recognition is based on the relevant data obtained from research achievements of Geo-dynamic Division, With the help of spatial data management, the Neuron Network and Cluster algorithm are applied to predict the danger probability of coal and gas outburst in each cell of coal mining district. So a coal-mining district can be divided into three areas： dangerous area, minatory area, and safe area. This achievement has been successfully applied for regional prediction of coal and gas outburst in Hualnan mining area in China.     

Advances in Educational Data Mining Models and the Application of Its Algorithms

In order to find an effective way to improve the quality of school management, finding valuable information from students’ original data and providing feedback for student management are necessary. Firstly, some new and successful educational data mining models were analyzed and compared. These models have better performance than traditional models (such as Knowledge Tracing Model) in efficiency, comprehensiveness, ease of use, stability and so on. Then, the neural network algorithm was conducted to explore the feasibility of the application of educational data mining in student management, and the results show that it has enough predictive accuracy and reliability to be put into practice. In the end, the possibility and prospect of the application of educational data mining in teaching management system for university students was assessed.     

基于决策树的就业数据挖掘

针对学生就业问题，给出了就业数据挖掘模型．决策树方法是数据挖掘中非常有效的分类方法，根据就业数据特点，采用了C4．5决策树算法．C4．5算法是决策树核心算法ID3的改进算法，它构造简单，速度较快，容易实现．模型对就业数据预处理，选取决策属性，实现挖掘算法并抽取规则知识，由规则知识指出哪些决策属性决定了就业单位的类别，挖掘结果表明，该算法能够正确将就业数据分类，并得到若干有价值的结论，供决策分析。     

基于数据挖掘的决策树算法分析

决策树是数据挖掘技术中的重要方法,主要用于分类和预测.本文介绍了决策树算法中应用最广泛的ID3算法和C4.5算法,阐述了两种算法的主要思想,说明了构造决策树的基本步骤,对两种算法进行了分析和比较.     

数据挖掘网格中决策树并行算法设计及性能分析

提出了C4.5决策树算法的一种并行算法,使传统的串行分类算法能在多台PC机和服务器组成的数据挖掘网格上并行数据挖掘. 采用数据纵横剖分,结合递归过程的并行化,实现了可扩展的高性能并行计算,解决了处理海量数据时没有较好并行分类算法的问题. 并给出了指导该并行算法高效计算的方法. 数据运行试验和算法分析表明,该并行算法的性能受多个因素影响,并具有高效的并行效率计算加速比.     

基于能量特征估计的电能质量扰动消噪方法研究

电能质量扰动识别过程中噪声的存在会增加误判,为了提高分类的正确率,消噪对于电能质量扰动识别是一项非常重要的工作。论文应用Daubechies小波时频分解的噪声能量保持特性来估计扰动信号中不同分解尺度上的噪声能量,从而由含噪声信号能量分布和所估计的噪声能量确定实际扰动信号的能量特征,完成了消噪,对消噪处理后电能质量扰动信号应用数据挖掘中的决策树算法进行识别。仿真计算表明,该消噪方法能提高识别精度,是一种非常有效的电能质量扰动信号消噪方法。     

C4.5算法在城建档案催交管理中的应用

决策树作为一种数据挖掘分类算法,不仅可以根据现存的数据库信息得出分类规则,而且可以通过友好的图形界面进行结果展示.城建档案馆在档案催交管理工作中,收集了大量的数字化信息,文中运用 C4.5 算法针对催交信息构造决策树,并利用生成的决策树辅助城建档案馆进行档案的催交管理.     

基于S变换和决策树算法的电能质量扰动识别

提出一种基于S变换和数据挖掘中决策树算法的电能质量扰动识别的方法.该方法首先用S变换对电能质量扰动波形进行时频分析,并使用统计方法提取相关特征量,然后用决策树算法对提取的特征量样本进行分类,并获得明确的分支规则.仿真结果表明,该方案正确率高,抗噪声能力强,训练样本少,响应速度快.     

一种基于属性频率划分的决策树算法

决策树是数据挖掘任务中分类的常用方法。在构造决策树的过程中,节点划分属性选择的标准直接影响决策树分类的效果。基于粗糙集的属性频率函数等方法度量属性重要性的概念,将其用于分枝划分属性的选择,提出一种决策树学习算法。该方法仅利用区分矩阵就可以计算出属性的出现频率函数值,计算简单。实验结果表明,用该方法构造的决策树与传统的基于信息熵方法构造的决策树相比较,结构简单,且能有效提高分类效果。     

基于规则的遥感影像分类方法研究——以黄山市为例

介绍了决策树C4.5算法,并利用该算法实现了对遥感数据规则的挖掘,在此基础上设计并实现了针对于C4.5规则的编辑器,通过该编辑器能够实现对规则的编辑与管理。规则编辑器的设计与实现,为在分类过程中人工的干预提供了可能。将人工干预与基于数学理论规则的自动提取相结合,尤其是在地形较复杂的地区,将有利于分类精度的提高。利用黄山市Land Sat TM影像,进行了基于C4.5算法自动提取规则的遥感影像分类实验。实验结果表明,利用C4.5算法提取的分类规则准确率高,利用提取的分类规则进行的遥感影像分类效果较好。     

基于C4.5算法的健身俱乐部会员数据挖掘研究

以已投入使用的健身俱乐部管理系统为背景,提出了用C4.5决策树分类算法对健身记录进行数据挖掘。通过该方法找出俱乐部在有效期内的会员的年龄段、性别、会员卡类型和参与健身时间段的规律,提取特定时间段内参与健身的会员特征。实验结果表明:将该分类规则应用到会员管理系统中,可以辅助健身俱乐部的管理者有针对性地制定营销方案。     

数据挖掘技术在CRM中的应用研究

介绍了数据挖掘技术和CRM的基本概念,分析了数据挖掘技术在CRM中的应用,并着重对决策树技术进行了研究。针对SPRINT算法在数值型属性分裂过程中计算量大、效率低的缺点,提出了一种新的数值型属性的分裂方法,并对SPRINT算法进行了改进,并用改进的SPRINT算法构建了客户信用卡的分类模型。