二维模糊集与随机集落影

本文以不同于常规的方法定义了二维随机集,及其联合分布、联合落影测度、联合落影函数,给出了成为二维随机集的充要条件。讨论了二维模糊集与随机集落影的关系。     

Effect of Azospirillum brasilense inoculation on rhizobacterial communities analyzed by denaturing gradient gel electrophoresis and automated ribosomal intergenic spacer analysis

Nucleic acid-based techniques allow the exploration of microbial communities in the environments such as the rhizosphere. Azospirillumbrasilense, a plant growth promoting rhizobacterium (PGPR), causes morphological changes in the plant root system. These changes in root physiology may indirectly affect the microbial diversity of the rhizosphere. In this study, the changes in the rhizobacterial structure following A. brasilense inoculation of maize (Zea mays) plants was examined by PCR-denaturating gradient gel electrophoresis (DGGE) and automated ribosomal intergenic spacer analysis (ARISA), using two universal primers sets for the 16S rRNA gene, and an intergenic 16S-23S rDNA primer set, respectively. Similar results were obtained when using either ARISA or DGGE performed with these different primer sets, and analyzed by different statistical methods: no prominent effect of A. brasilense inoculation was observed on the bacterial communities of plant roots grown in two different soils and in different growth systems. In contrast, plant age caused significant shifts in the bacterial populations.     

Dynamical niche sets-based cooperativeevolutionary algorithm for job shop scheduling

Job shop scheduling is an important part of manufacturing process. Pre-mature, local optimal solution or low convergence rate of Genetic Algorithm may come across in solving scheduling problem in multi-Job shop. To avoid those unhealthy tendencies, this paper introduced the Dynamical Niche Sets-based Cooperative Evolutionary Algorithm model.In the Algorithm, Process-based chromosome encoding method was used; the crossover operator and mutation operator were designed to adjust operation sequences and choose process routes. The cooperative scheduling problem in multi-Job shop was solved using dynamical niche sets technology and multi-population cooperative evolution method. The Experiment result shows the proposed method has good advantages.     

主客观结合确定网络课程评价指标权重

通常，指标权重的确定方法是根据决策者的先验知识来确定的，其中比较常用的方法有德尔菲法及层次分析法，这些方法受决策者主观因素的影响太大。本文将粗集理论中属性重要度的概念引入网络课程评价中的权重确定过程，提出将决策者先验知识给定的权重同粗集理论确定的属性重要度结合起来作为最终确定网络课程评价指标权重的方法。     

基于粗神经网络的旅游需求预测研究

旅游需求的合理预测是旅游业持续健康的重要前提之一,根据粗糙集理论提出一种旅游需求预测的粗神经网络模型,该模型通过消除噪声提高神经网络预测的精度,有效地克服神经网络算法中知识获取以及推理过程的黑箱性弱点。通过对云南近年入境旅游发展趋势的预测实例分析,验证模型的有效性。     

农业信息数据挖掘的粗糙集方法研究与实例

我国是个农业大国,农业数据量极大,手工处理工作量大且不准确.本文研究了将粗糙集理论运用于农业数据挖掘的计算机处理方法,提出了基于粗糙集理论的农业数据挖掘的基本算法和模型,并通过实例对其进行了验证.     

基于粗糙集理论的河南省农民收入预测研究

采用理论和实证分析相结合的方法,从自身、全国平均水平、中部相邻省份3个层次对比研究河南省农民收入变动情况。结合河南省的实际数据,运用粗糙集理论计算出各因素对河南省农民收入的影响系数,以及若干条确定性预测规则。最后对预测结果进行了分析。     

聚类分析在西瓜抗枯萎病品系筛选中的应用

对28个西瓜品种进行抗枯萎病筛选．苗期用特定浓度接种，每隔7d调查其感病情况及植株生长情况．株高、茎粗、叶片数等生长势指标与发病率均呈极显著负相关，相关系数分别为－0．5480、－0．6052、－0．8114．利用聚类分析方法统计，把28个西瓜品种按其抗性表现归类，从而得出抗病水平不同的各组西瓜品种，并筛选出一组抗性水平相当高的品种．为选育高抗西瓜品种以及研究各西瓜品种的抗病性关系提供了一种简捷的科学方法．     

测度贴近度及其集值统计估值

在定义模糊集上的测度的基础上，通过对随机集测度均值的讨论，定义了一个无穷论域上的贴近度－测度贴近度，为贴近度的计算提供了新方法，充实了贴近度的定义开式。又根据模糊集与随机集落影响的理论，给出了有严格理论基础和应用性较强的对贴近度进行集值统计的方法，并且证明了这种估计的一致性。