Mixture equations of state: Composition dependence

This paper discusses theoretical models for the composition dependence of equations of state and compares the quality of predictions against experimental thermodynamic property data. The mean density approximation (MDA) and the van der Waals one-fluid (VDW1) model are compared with hybrid mixing rules (HMR), in which rigorous composition dependence is used for the second and third virial coefficients and the conformai solution model is used for equation-of-state density terms beyond the third virial term. It is found that when values of unity are used for all binary and three-body unlike interaction parameters, calculated densities for methane-normal heptane mixtures have average absolute deviations of 3.54% for MDA, 4.04% for VDW1, and 2.59% for HMR. When vapor-liquid equilibrium calculations were performed for the methane-normal heptane system, average absolute deviations of calculatedK values from experimental values were 16.7% for methane and 36.4% for normal heptane using HMR, whereas when conformal solution model (CSM) mixing rules were used, the results were 34.8% for methane and 66.7% for normal heptane. When the binary interaction parameter for the characterization of interaction energies is determined, it is found to be less sensitive to state conditions in the case of HMR than either MDA or VDW1. These preliminary results suggest the potential of mixture equation-of-state methods which utilize rigorous composition dependence for the lower-order virial coefficients.Paper presented at the Second U.S.-Japan Joint Seminar on Thermophysical Properties, June 23, 1988, Gaithersburg, Maryland, U.S.A.     

Genome-Wide Association Study Identifies a Rice Panicle Blast Resistance Gene,Pb2, Encoding NLR Protein

Rice blast is one of the main diseases in rice and can occur in different rice growth stages. Due to the complicated procedure of panicle blast identification and instability of panicle blast infection influenced by the environment, most cloned rice resistance genes are associated with leaf blast. In this study, a rice panicle blast resistance gene, Pb2, was identified by genome-wide association mapping based on the panicle blast resistance phenotypes of 230 Rice Diversity Panel 1 (RDP1) accessions with 700,000 single-nucleotide polymorphism (SNP) markers. A genome-wide association study identified 18 panicle blast resistance loci (PBRL) within two years, including 9 reported loci and 2 repeated loci (PBRL2 and PBRL13, PBRL10 and PBRL18). Among them, the repeated locus (PBRL10 and PBRL18) was located in chromosome 11. By haplotype and expression analysis, one of the Nucleotide-binding domain and Leucine-rich Repeat (NLR) Pb2 genes was highly conserved in multiple resistant rice cultivars, and its expression was significantly upregulated after rice blast infection. Pb2 encodes a typical NBS-LRR protein with NB-ARC domain and LRR domain. Compared with wild type plants, the transgenic rice of Pb2 showed enhanced resistance to panicle and leaf blast with reduced lesion number. Subcellular localization of Pb2 showed that it is located on plasma membrane, and GUS tissue-staining observation found that Pb2 is highly expressed in grains, leaf tips and stem nodes. The Pb2 transgenic plants showed no difference in agronomic traits with wild type plants. It indicated that Pb2 could be useful for breeding of rice blast resistance.     

基于属性变化的增量关联规则挖掘

关联规则的挖掘是数据挖掘研究中的一个重要课题,目前已经提出了许多用于发现海量事务库中关联规则的算法以及更新已经发现的关联规则的算法。但是在关联规则的更新算法中,都是基于支持度变化和事务库变化的研究,目前没有人研究当事务库中的属性发生变化时,如何高效地更新关联规则的问题。针对这种情况,提出了三种基于属性变化的增量关联规则挖掘算法ACA+（Attribute Change Algorithm）和ACA-（ACA1-）,从而解决了该问题。     

基于Vague集的Apriori算法的改进

当前研究的关联规则挖掘隐含着一个假定：待挖掘数据库中每条记录的全部属性都是完整的。然而现实中得到的数据可能是不完整的,在这种情况下将无法使用传统的算法。为了解决这个问题,引入Vague集并给出基于Vague集的支持度和置信度定义,将经典的Apriori算法推广到Vague集上。最后通过一个例子来验证算法的有效性。     

基于Rough集方法的RS-MAXNET网络分类器模型

Rough集（RS）理论中如何补偿或拓展Rs约简的局部决策规则使之更加接近全局规则是一个重要的议题。本文提出RS与MAXNET网络结合拓展建立盼MAXNET全局网络分类器模型。基于上述模型进行的脱机手写模式识别算法说明了其有效性。     

求解无等待流水调度的基于禁忌搜索的混合算法

本文提出了解决最小完工时间的无等待流水调度问题的基于禁忌搜索的混合算法。算法结合了调度规则和禁忌搜索算法的优点,首先利用调度规则构造较好的初始解,既可以加快禁忌搜索算法的收敛速度,也可以降低整个算法的运算量,使算法有更好的工程实用性;然后使用变邻域结构的禁忌搜索算法改进当前解。在保持可达性的基础上,该算法缩小了邻域规模和减少了计算时间。数值仿真实验表明,该算法是有效的。     

新属性重要性的规则提取方法

针对ID3算法倾向于取值较多的属性问题,利用粗糙集理论,提出一种属性选择的新度量方法。该方法借助信息熵的性质充分考查规则集整体的确定性和不确定性问题。分析规则的三个度量标准：支持度、置信度、覆盖度,对其进行量化,并提出一规则提取算法。实例表明,该算法提高了决策分类的精度,进而能提取可靠有效的规则。     

频繁模式挖掘进展及典型应用

对近年来频繁模式的挖掘进行了总结。首先对有代表性的挖掘算法从算法思想、关键技术、算法的优缺点进行了分析概括,此后列举了一些典型频繁模式及关联规则的领域应用。综述内容的选择主要基于某一个研究后续被关注程度,组织过程中力争将相关研究进行归类,以给出其发展概貌。上述工作可以为频繁模式挖掘及关联规则的研究提供有益的参考。     

The Skeptical Explorer: A Multiple-Hypothesis Approach to Visual Modeling and Exploration

The primary intent of this work is to present a method for sequentially associating three-dimensional surface measurements acquired by an autonomous exploration agent with models that describe those surfaces. Traditional multiple-viewpoint registration approaches are concerned only with finding the transformation that maps data points to a chosen global frame. Given a parts-based object representation, and assuming that the view correspondence can be found, the problem of associating the registered data with the correct part models still needs to be solved. While traditional approaches are content to group segmented data sets that geometrically overlap one another with the same part, there are cases where this causes ambiguous situations. This paper addresses the model-data association problem as it applies to three-dimensional dynamic object modeling. By tracking the state of part models across subsequent views, we wish to identify possible events that explain model-data association ambiguities and represent them in a Bayesian framework. The model-data association problem is therefore relaxed to allow multiple interpretations of the object's structure, each being assigned a probability. Rather than making a decision at every iteration about an ambiguous mapping, we look to the future for the information needed to disambiguate it. Experimental results are presented to illustrate the effectiveness of the approach.     

基于免疫克隆选择算法的神经网络规则抽取

神经网络的不可解释性一直是限制其发展的固有缺陷,该文从神经网络的功能性观点出发,提出基于免疫克隆选择算法的神经网络规则抽取方法。将免疫克隆策略用于神经网络的规则抽取中,对已经训练好的神经网络隐层神经元输出值进行聚类,缩小搜索空间,抽取出理解性好、简洁的符号规则。该方法不依赖于具体的网络结构和训练算法,可以方便地应用于各种分类器型神经网络。实验结果表明该方法的实用性和可行性。