基于FP-树的关联规则挖掘算法浅谈

随着数据库技术的发展,高效的数据挖掘算法有助于人们重新认识数据、理解数据。基于FP-树的关联规则挖掘算法FP-growth是当前应用最广的一种挖掘频繁项目集的算法。本文简要描述了该算法的几种主要发展方向。     

基于压缩事务矩阵相乘的Apriori改进算法

Apriori算法是当前使用最广泛的关联规则挖掘方法中最为经典的算法之一;但是该算法需要反复的扫描数据库,在I/O上花消很大,并且在得到频繁-2项集的过程中会产生庞大的候选-2项集,其次在筛选得到频繁-k项集时,并没排除那些不应该参组合的元素,而导致该算法效率很低,针对上面影响计算效率的三个方面提出基于压缩事务矩阵相乘得到频繁项目集的算法,只需一次扫描数据库,经过压缩处理产生产生事务矩阵,通过矩阵间运算得到频繁项目集,有效提高了关联规则的挖掘效率。     

一种频繁项集挖掘改进算法

利用Apriori算法和FP-growth算法挖掘密集型数据集的全部频繁项集代价高昂,针对该问题提出一种基于链表数组的关联规则挖掘算法,该方法使用链表数组为每个项目建立事务链表,只需要扫描数据库1次,就能够快速得到每个候选项的支持度,从而有效的发现频繁项集。通过与经典算法分析对比表明,该算法具有较快的挖掘速度。     

一种基于目标属性的关联规则挖掘算法的研究

提出一种基于目标属性的关联规则挖掘算法,该算法对于不同目标属性的关联规则挖掘是相互独立的,虽然会产生重复工作,但是在并行计算以后效率得到了大大的提高.经过实验分析,对于同样的问题,采用该算法并引入并行计算后,效率要比采用间接挖掘方式的Apriori算法高得多.     

Apriori算法在研究生成绩中的应用

介绍了关联规则的基本概念，利用Apriori算法得出频繁项目集，结合最小支持度和最小之置信度，确定各门课程之间的联系，为研究生的选课提供一定的参考。     

数据挖掘中关联规则的频繁项集研究及算法设计

关联规则挖掘其主要研究目的是从大型数据集中发现隐藏的、有趣的、属性间存在的规律与数据间的联系。关联规则挖掘算法主要目的是从事务数据集项间挖掘出有意义的关联关系。Apriori算法是关联规则挖掘算法中最经典的方法。由Apriori算法产生的候选项集仍是巨量的。通过对Apriori算法中的候选项集支持频度的深入研究总结五条规律,并将这五条规律应用到Apriori算法中。     

一种基于fp-tree的Apriori算法改进研究

提出了一种改进的基于fp-tree的Apriori算法.该算法先用尾元将fp-tree分区,生成数据量更小的子数据集,再动态删除冗余数据将子数据集的数据进一步压缩,最后通过扫描子数据集进行支持数统计,从而快速挖掘.实验结果表明,在对含有大量高维度数频繁项集的数据集进行挖掘时,这个改进算法的挖掘速度较快.     

一种高效挖掘生物网络闭合频繁子图的算法

针对生物网络中频繁子图的挖掘问题,提出了一种基于FP-树结构的MaxFP算法.此算法以代谢路径作为研究对象,在适合于生物网络图简化模型的基础上,采用一种不产生候选集的改进FP-growth算法挖掘生物网络中的闭合频繁子图.此算法考虑了基于频繁项目集的算法应用于网络的缺陷,根据生物网络的特点对FP-growth算法进行了改进.实验证明,提出的MaxFP算法比基于Apriori的频繁模式挖掘算法运行速度快,不仅能挖掘出最大的频繁子图,且能找到更多具有生物意义的频繁子图.     

基于关联规则的恒星光谱数据相关性分析

利用关联规则作为恒星光谱数据相关性分析方法,以VC 和Oracle9i作为开发工具,设计并以程序实现了基于关联规则的恒星光谱数据相关性分析系统,给出了其软件体系结构和模块功能,并对恒星光谱数据的预处理、取样技术、背景知识模板、频繁模式集挖掘算法等关键技术进行了详细描述.最后,通过系统运行结果表明,利用关联规则来描述、分析光谱数据特征和物理化学性质之间存在的相关性是可行的和有价值的.     

分布式数据库中关联规则挖掘的研究

在数据挖掘研究中,关联规则挖掘作为数据挖掘研究中的一个重要部分,引起越来越多的关注。因此,主要研究关联规则挖掘,首先介绍关联规则挖掘的一些基础知识、概念描述等,然后对关联规则挖掘的常用算法进行分类探讨,最后分析其中的几种典型算法。     

Solution to industrial optimization problems through differential evolution variants

Optimization forms the core of several industrial problems and demands suitable methods of solution for determining the best possible solution. However, it is often observed that the classical optimization algorithms available in literature may not be applicable in real-life scenarios where the nature of the problems is quite complex. Under such circumstances, nature-inspired algorithms (NIA) can be beneficial because of their generic nature. This is shown in the present study by obtaining the optimal solution of two fundamental industrial problems, viz. optimizing and controlling the level of noise in an industry having multiple sources of noise; and optimizing the total production cost in a machining process through different NIA including two newly proposed differential evolution (DE) variants. Comparison is also done through classical methods. Both the problems are nonlinear in nature where the complexity increases by increasing the number of variables. The first problem taken is unconstrained in nature, while the second problem is constrained. A thorough comparison of all the methods is done through various performance measures and it is observed that the proposed DE variants form an attractive alternative for dealing with such problems.     

A note on approximation of distribution function for the number of innovation activities

This note attempts to approximate the distribution function for the number of innovation activities (NIA) in the manufacturing sector using the dataset of 2002 Korean Innovation Survey. The mixture model applied here can easily capture the bimodality feature of the NIA distribution and provide some useful information such as the mean of NIA and the effect of a firm’s characteristic on whether the firm will undertake innovation activity.     

基于周期采样的数据流频繁项集挖掘算法研究

针对用于数据流频繁项集挖掘的现有方法存在引入过多次频繁项集以及时空性能与输出精度较低的问题,利用Chebyshev不等式,构造了项集频度周期采样的概率误差边界,给出了动态检测项集支持度变化方法.提出了一种基于周期采样的数据流频繁项集挖掘算法FI-PS,该算法通过跟踪项集支持度变化确定项集支持度的稳定性,并以此作为调整窗口大小以及采样周期的依据,从而以一个较大的概率保证项集支持度误差有上界.理论分析及实验证明该算法有效,在保证挖掘结果准确度相对较好的条件下,可获得较优执行性能.     

基于模糊支持度的分布式多传感器加权融合算法

针对目前分布式多传感器数据融合算法中鲁棒性和实时性问题,基于充分利用多传感器测量数据中互补和冗余信息的思想,通过局部状态估计精度间支持度函数的建立和支持度矩阵的求解,合理地分配各局部航迹估计在融合中心的权重,进而提出了一种基于模糊支持度的分布式多传感器加权融合算法.最后,通过蒙特卡罗仿真验证了该算法的有效性.     

Mining highly correlated least association rules using scalable trie-based algorithm

Development of least association rules (ARs) mining algorithms is one of the more challenging areas in data mining. Exclusive measurements, complexity and excessive computational cost are the main obstacles as compared to frequent pattern mining. Indeed, most previous studies still use the Apriori-like algorithms. To address this issue, this article proposes a new correlation measurement called definite factor (DF) and a scalable trie-based algorithm named significant least pattern growth (SLP-Growth). This algorithm generates the least patterns based on interval support and finally determines it significances using DF. Experiments with the real datasets show that the SLP-Growth can discover highly positive correlated and significant of least ARs. Indeed, it also outperforms the fast frequent pattern-Growth algorithm up to two times, thus verifying its efficiency.     

Comparison of the Non-Interaction and Differential Schemes in predicting the Effective Elastic Properties of Fractured Media

The non-interaction approximation (NIA) formulated in compliances and the differential effective media (DEM) schemes are believed to be the most accurate theories for predicting the effective elasticity of fractured solids. While their predictions are always plausible, the DEM yields consistently softer effective properties than does the NIA. Here I compare these two theories with the finite element (FE) modeling for arrays of randomly located, parallel, penny-shaped cracks. I perform FE simulations by applying the homogeneous strain and homogeneous stress boundary conditions that establish the upper and lower bounds for the effective stiffness tensor. These numerically derived bounds demonstrate that the NIA is more accurate than the DEM.     

适合于地震工程信号分析的快速小波变换法研究

本文简要分析了目前普遍采用的小波变换法,指出其尚不能满足地震工程信号分析的要求,由此提出一种直接变换法的改进方法,经详细的考查,证明该方法精度高、速度快,能够满足地震工程信号分析的要求。     

静力荷载作用下结构参数识别及状态评估的统计分析

基于静态结构识别方法,提出了服役结构损伤探测及状态评估的概率分析方法。首先通过有限元法把结构离散成用基本参数及核心矩阵表示的分析模型,根据两种分析模型和实际结构间的误差定义,用 Gauss-Newton法推导了在不完全测量情况下两种参数识别方法。在设计的测量情况下,进一步采用 Monte Carlo 法模拟测量数据,详细分析和比较这两种算法在有测量误差时的有效性和稳定性问题。第一种算法中的敏感矩阵不受测量误差的影响而具有比较好的性能,在确定了其识别结果的概率特性后,引入假设试验,探测和评估损伤的位置及程度。数字模拟显示了这一方法的有效性和准确性。     

Radial basis function method for 1-D and 2-D groundwater contaminant transport modeling

In this paper we develop a meshless method for modeling groundwater contaminant transport. The algorithm uses collocation method with radial basis functions. Numerical results are presented for several cases: pure diffusion; advection and dispersion for continuous source; advection and dispersion for instantaneous source; advection and dispersion for patch-source. The results show that the method is very simple and accurate.Jichun Li likes to thank the UNLV Research Office for the NIA grant support for this project. This work is also partially supported by Nevada EPSCoR. The authors are very grateful to the reviewers for their valuable comments on improving this paper.     

Current advances and future trends in X-ray digital detectors formedical applications

Despite the progress made in imaging modalities based on nonionizing radiation such as ultrasound and magnetic resonance imaging, X-ray is still the primary imaging method in most radiology departments. In recent years, digital electronics and manufacturing methods have led to many approaches for the design and construction of X-ray detectors. The ideal medical X-ray system is a digital, flat-panel able to perform all current radiological modalities at reduced dose. It would, immediately after the patient's X-ray exposure, provide a high quality radiograph on a video monitor. It would also be usable for real-time imaging (e.g., digital cine loops and fluoroscopy). The physical form would be similar to a film/screen cassette (except for the addition of an umbilical cord) which could easily fit into the Bucky tray of existing X-ray rooms. There are two main approaches both based on active matrix technology developed for lap-top displays. The first, called the direct method employs a photoconductive layer of amorphous selenium (a-Se) to detect X-rays. The charge image formed on the surface of the a-Se is digitally read out in situ using a two-dimensional array of thin-film transistors (TFT's), i.e., an active matrix. The second flat panel method is called the indirect method and uses an active matrix to readout a phosphor layer. Phosphors give off light on interaction with X-rays and this is sensed by photo diodes on the active matrix. The relative practicality of the direct and indirect active matrix readout systems is an area of current investigation