一种集成数据挖掘的自动视频分类方法

针对自动视频分类工作中分类预测精度低的问题，提出了一种集成数据挖掘技术的自动视频分类方法。首先进行视频分割，形成了一个视频属性数据库；然后分别使用决策树、分类关联规则等技术对视频属性数据库进行数据挖掘，提取出决策树分类规则集和分类关联规则集；最后利用一个规则集的合并裁减算法来合并这两个分类预测规则集，形成最终的具有更高精度的视频分类规则集。通过实验验证了决策树分类预测规则和分类关联规则具有分类预测的一致性；同时实验表明，使用合并后的规则集比单独使用一个规则集来预测视频具有更高的预测准确率。     

一种量化关联规则挖掘算法

提出了一种新的挖掘量化关联规则的方法。该方法使用聚类算法把数据库中的交易记录分成若干个簇，把簇投影到数值型属性所在的域，形成重叠的、有意义的区间。实验结果显示，这种方法能够有效地挖掘量化关联规则，并且能够发现以前的算法可能遗漏的重要的规则。     

一种期货市场关联交易行为检测的聚类方法

期货市场中的关联交易是金融监管的核心问题之一。批量交易数据条件下,关联交易行为发现的难点是如何在有限时间和空间条件下快速发现违规行为。针对该问题,提出一种将交易数据的稀疏相似性矩阵映射到十字链表的聚类算法。通过在真实交易数据下的实验分析,验证该算法能够有效解决关联交易行为的发现问题;且减少了扫描数据库的次数,降低了传统算法的时间复杂度和空间复杂度。     

基于时空快照数据库的时间序列预测

时空数据库是在空间数据库的基础上引入了时间维,时空数据模型和时空变化分析是GIS领域当前研究热点之一。提出一种在时空快照数据中预测时间序列发展和关联规则发现的方法。首先采用基态修正模型表达时空数据,从中提取出时空快照序列,将时空快照序列聚类为几个簇,再在簇内挖掘关联规则。将该方法应用于实验数据,结果证明这种方法能够有效地从时空快照数据中发现时空序列的发展趋势。     

一个数据挖掘工具的设计与实现

数据挖掘是一个利用各种分析工具在海量数据中发现模型和数据间关系的过程，这些模型和关系可以用来做出预测，该文介绍了一人数据挖掘工具的设计，以Apriori算法为核心，实现了数据挖掘中基于数据库的几种常用挖掘方法，包括基于关系数据库的数据挖掘，不完整数据库中的数据挖掘和根据兴趣度测量来挖掘感兴趣知识的异常关联规则挖掘。     

时序数据挖掘方法研究

数据库知识发现的基本思想就是从数据中抽取有价值的信息，其目的是帮助决策者寻找数据间潜在的关联，发现被忽略的要素，而这些信息对预测趋势和决策行为也许是十分有帮助的。时序数据挖掘是数据库知识发现研究中的重要分支之一。趋势分析与相似搜索是时序数据挖掘的主要技术与方法。通过趋势分析，可以制定出比较合理的长期或短期预测，从而为科学决策提供有效的依据；而在相似搜索中，采用了模糊匹配技术，符合人脑思维特性，因而更合理有效。     

一种基于关联规则的网络软件缺陷预测方法

针对网络软件缺陷预测中缺陷之间的显性关联关系和隐性关联关系,为提高网络软件缺陷预测的精确度和效率,基于现有的软件缺陷数据集,提出一种基于关联规则的网络软件缺陷预测方法。首先采用随机方法从美国国家航空航天局NASA的软件缺陷数据库中提取用于分类和测试的数据集,利用关联规则方法中Apriori算法对数据集进行关联规则生成和关联分类器的构建,并与BP神经网络方法的预测结果进行对比。结果表明,基于关联规则方法能够在小样本数据集中提高网络软件缺陷预测的精确度和有效性。     

基于中药专利数据集的关联规则发现算法*

指出关联规则在中药数据分析中的难点,据此提出了一种改进的Apriori算法--Apriori 算法;最后,以治疗感冒的中药专利数据集为测试数据,进一步验证算法的有效性和实用性.结果表明,此算法能够有效地从治疗感冒的专利数据库中发现布尔型与数值型关联规则,为开发新的感冒中药提供配伍依据.     

基于概率统计的数据挖掘预测算法

预测是数据挖掘中的重要功能,讨论基于概率统计的贝叶斯数学模型,利用已知数据的统计特性,通过相关性分析,对数据未知的某些属性进行预测和解释.     

基于RBF网络的固相质量流量检测

气固两相流流动的复杂性和多样性直接影响了固相质量流量的测量精度,严重限制了其在工业生产中的应用。在双弯管法检测固相质量流量原理的基础上,利用人工神经网络优良的非线性映射能力,以两弯管处的压力差为输入,建立了一种基于径向基(RBF)网络的软测量模型,实现了对固相质量流量的在线测量。实验结果表明:该模型测量误差在3%以内,能够有效解决固相质量流量在线检测的问题,并且具有测量快速、精度高的优点。     

循环流化床气固两相流动模拟

循环流化床已被广泛用于能源、化工、环保等工业领域,但由于流化床内两相流动、传热及化学反应的物理机理和作用规律复杂性,目前为止对流化床的认识还远远不能令人满意,因此了解流化床内流动机理对循环流化床的设计和运行有深远的指导意义.针对循环流化床燃烧技术,建立了描述其炉内气固运动特性的三维数学模型,用Fluent软件作计算工具,利用欧拉双流体模型(EULERIAN-EULERIAN)对流化床内的颗粒浓度分布、颗粒速度分布和床内压力分布等进行了三维数值模拟,计算结果表明:在床内固体颗粒浓度中心区域低、近壁面高的环核结构,固体颗粒在横截面上存在由核心区向环形区的内循环运动,在相同流化风速度下,沿床高压降随循环物料的增加而变大,在相同的物料循环量,沿床高压降随着流化风速的增加而减小.     

槽式孔板的气液两相流测量特性

为了准确计量气液两相流量,选用槽式孔板作为一次传感元件、以空气/水为测量介质,进行了一系列的实验.介于两相流动的随机波动特征,对气液两相流流过槽式孔板时产生的差压信号进行了处理,提取出能够反映气液流量变化的几个统计特征量,总结了差压信号的平均值、标准差、自相关最大值和(0,4)Hz频段功率随气液流量变化而变化的规律,为两相流流量计的开发奠定了基础.     

循环流化床锅炉整体自动建模系统的数字模型

该文针对循环流化床锅炉建模问题，设计了新的整体搭建模型步骤，并在先进的建模工具－计算机代数系统的基础上开发了基于符号计算的循环流化床锅炉自动建模系统。与国外同类产品相比，本系统的推演更高效、更直观，灵活性更强。实践证明，带有自动建模系统的循环流化床锅炉整体模型能够节约一半左右的模型开发时间，使研究人员从繁琐的手工推导和编程工作中解脱出来，具有很强的实用性，该文建立整体模型的思路和自动建模系统还可以推广到鼓泡流化床，增压流化床、气化炉等一类热能动力系统的建模中，为其它对象的模型研究奠定了基础。     

Graph directed locking

A non-two-phase database concurrency control technique is introduced. The technique is deadlock-free, places no restrictions on the structure of the data, never requires data to be reread, never forces a transaction to be rolled back in order to achieve serializability, applies a type of lock conversion, and allows items to be released to subsequent transactions as soon as possible. The method introduced, database flow graph locking (FGL), uses a directed acyclic graph to direct the migration of locks between transactions. Unlike many previous non-two-phase methods, the database need not be structured in any specific fashion. The effect of these changes is that, with the same serializable schedule, FGL obtains a higher degree of concurrency than two-phase locking (2PL). Overhead requirements for database flow graph locking are comparable to those for two-phase locking, with 2PL being better in low conflict situations and FGL better in high conflict     

循环流化床锅炉床温调整及自动控制策略探讨

循环流化床锅炉运行中,床温是最重要的监视及控制参数之一,目前国内大部分电厂运行中存在的床温波动较大,无法投入自动控制等问题,本文对锅炉启动及锅炉及正常运行过程中手动床温调整方法进行了研究,并提出了几点床温自动控制策略的改进建议。     

Locking performance in a shared nothing parallel database machine

A quantitative performance study of two-phase locking in a parallel database machine using a simulation-based methodology is described. The DBSIM simulation methodology uses models at two levels: a Petri net model at the higher level and a queuing network model at the lower level. The Petri net model captures the characteristics of parallelism and synchronization at the workload level, while the queuing network model captures queuing aspects of the system at the physical resource level. Transactions in a workload are specified using a performance-oriented specification language based on the transaction component graph, a data flow graph with database operators. The transaction specifications are translated into Petri net representations to derive the simulation experiments. The workload is a transaction taken from an order-entry application. A shared-nothing parallel machine architecture is assumed. Results of analysis of a two-phase locking strategy with machine sizes ranging from 4 to 256 processors are presented     

An Object Flow Computer for Database Applications: Design and Performance Evaluation

In the past two decades, a considerable amount of research exists which uses hardware, firmware, and novel architectures to achieve the needed efficiency in implementing database management functions. However, most of the past efforts have been directed toward developing database computers for supporting a relatively primitive data model, namely, the relational model. This paper presents the design and evaluation of an Object Flow Computer (OFC). OFC is designed to efficiently support the processing of object-oriented databases. OFC employs a vertically fragmented data storage structure and a two-phase parallel query processing strategy. A set of primitive operators is defined for OFC. Depending on the performance requirement, these operators can be implemented in software running on general-purpose processors or as functions in special-purpose coprocessors. A high-level database request can be decomposed into these primitive operators and executed in parallel. OFC combines a number of known database processing techniques such as query decomposition, pipelining mode of data processing, and data flow control strategy. The performance evaluation of the proposed two-phase query processing strategy and a comparison with the conventional query processing strategy are also presented.     

多模态动态核主成分分析的气液两相流状态监测

气液两相流流动过程作为一种非平稳过程,其状态的变化具有时变性、非线性、随机性等复杂流动过程的特点,其流动状态的实时监测对掌握其流动过程的产生、发展及转化,保障实际生产的安全稳定运行具有重要意义.特别是流动状态的过渡过程反映了流动状态的发展及演化,其流动结构非常复杂.针对气液两相流的3种典型流动状态及过渡转化过程,在多传...     

Behavior analysis of air bubbles in the oil lubricant of FDBs at low speed operating conditions

This paper investigated the behavior of the air bubbles and the air–oil interface of FDBs at low speed operating conditions by using a two-phase flow analysis of air and oil to study the expelling mechanism of air bubbles in the operation of FDBs. The two-phase flow of air and oil was analyzed using the three-dimensional Navier–Stokes equation and the volume of fluid method of multi-phase flow. The proposed numerical method was verified by the experimental results. The effects of the depth and area of the groove near the outlet on the behavior of the air bubbles and the air–oil interface were also studied. This research shows that the groove near the outlet is responsible for the concave shape of the air–oil interface as a result of surface tension, and that air bubbles are expelled into the outside air when an air bubble meets the concave air–oil interface.     

Spatially discontinuous ionization phenomenon in inhomogeneous soil

By using X-ray film imaging technology,a phenomenon of discontinuous ionization was observed in the inhomogeneous soil around the grounding electrode on which a surge voltage was applied.A simplified two-phase solid-gas model was built to study the electric field distribution in the soil to explain the discontinuous ionization phenomenon.Analysis showed the differences of the dielectric properties,the shapes and sizes of soil particles can cause discontinuous ionization in the soil.