混合语义相似度计算优化模糊查询的智能信息检索算法

针对搜索引擎领域存在的因查询条件模糊而严重影响信息检索性能的问题,提出了一种混合语义相似度优化模糊查询算法。首先,根据相似度计算概念提出了与其接近的新概念提取方法;然后,利用TF-IQF模型将链接划分成标记,并使用这些标记组成的集合表示出查询;最后,构建二分图识别出相关的查询,使用二分图计算出查询相似度。实验结果表明,相比点击文档、相关查询和反向查询三种算法,取得了更好的全查率、检索精度及F-度量。     

一种基于案例的多属性综合评价方法

案例推理是动态决策环境下求解不良结构问题常用的方法。把基于案例的推理方法运用于多属性综合评价中,扩展了传统数学模型的求解方式,详细叙述了基于案例的属性权重,建立算法和相似性度量算法,并将该方法应用于武器装备维修性设计方案权衡分析过程中。     

A similarity-based approach of Kuhn’s no-overlap principle and anomalies

In their book Cognitive Structure of Scientific Revolutions, Hanne Andersen, Peter Barker, and Xiang Chen reconstruct Kuhn’s account of conceptual structure and change, based on the dynamic frame model. I argue against their reconstruction of anomalies and of the no-overlap principle and propose a competing model, based on the similarity relation. First, I introduce the concept of psychological distance between objects, and then I show that the conceptual structure of a theory consists of a set of natural families, separated by a significant empty space. I argue that, in such a conceptual structure, the ES condition, according to which the distance between natural families should be greater than the distance between any two objects belonging to the same natural family, is satisfied. Anomalous objects lead to the violation of this condition. I argue that in a conceptual structure satisfying the ES condition, a similarity relation could be defined, so that natural families would be similarity classes, satisfying the no-overlap principle. In a structure not satisfying this principle, such similarity classes could not be delimited.     

一种基于互信息的词聚类算法

基于类的统计语言模型是解决统计模型数据稀疏问题的重要方法.传统的统计聚类方法基于贪婪原则,常以语料的似然函数或困惑度(perplexity)作为评价标准.这种传统的聚类方法的主要缺点是聚类速度慢,初值对结果影响大,易陷入局部最优.本文利用互信息定义了一种词相似度,在词相似度的基础上给出了词集合相似度的定义.基于相似度,提出了一种自下而上的分层聚类算法,这种方法不但能改善聚类效果,而且可根据不同的模型选择不同的相似度定义,因而提高聚类的使用效果.实验证明,该算法在计算复杂度和聚类效果上比传统的基于贪婪原则的统计聚类算法都有明显的改进.     

直觉模糊评价矩阵群决策

研究直觉模糊评价矩阵群决策问题.分别从相关性和相似性的角度提出与决策群体意见整体一致性程度最高的＂理想＂直觉模糊评价矩阵构造优化方法,利用该评价矩阵来评价专家的评判水平,说明本文提出的＂理想＂直觉模糊评价矩阵构造思想与现实中群决策的折衷思想是一致的.     

Vague集之间的相似度量分析

Vague集的相似度量是Vague集在各个应用领域中的关键技术．本文根据相似工程学原理，指出一个相似度量必须满足的约束条件，提出一种改进的Vague集相似度量公理化定义，然后引入一种新的Vague集相似度量方法，并证明它满足这些公理化条件，最后，用实例说明其应用以及该方法的有效性和直观性.     

一种改进的基于云计算的PageRank算法

PageRank算法是最为经典的Web结构挖掘算法,但是其存在主题漂移的问题,使得搜索结果中存在大量与查询主题无关的网页。在分析Page Rank算法的基础上,提出利用欧式距离计算主题相似度并融入传统的PageRank算法中,形成一个改进的网页排序算法,并把此算法应用到云计算环境中,研究MapReduce编程模型上的PageRank算法流程。     

视频人脸识别中基于聚类中心LLE的特征相似性融合方法

针对大部分现有视频人脸识别方法通常仅利用代表性范例或图像集而较少研究有效融合的问题,提出了一种基于聚类中心特征相似性融合方法。首先,使用局部线性嵌入从原始数据空间学习低维嵌入,并利用STHAC算法将投影划分为LLE特征空间聚类;然后,从基于局部外观的聚类中得到特征相似性,在贝叶斯最大后验概率分类框架中对范例点和聚类子空间进行相关相似性匹配;最后,借助于范例重要性概率完成人脸的识别。在视频人脸数据集CMU Mobo、Honda/UCSD和ChokePoint上的实验验证了所提方法的有效性,实验结果表明,相比几种传统的方法,所提方法取得了较高的识别精度和较低的计算复杂度。     

皖江城市带产业结构分析与协调发展研究

在构建产业同构和行业分工测评指标体系基础上,运用相似系数和专业化系数对皖江城市带产业整体及工业行业相似程度、专业化程度进行评价,明确皖江各市优势产业及相关行业,提出皖江城市带产业协调发展提升路径．     

The morals of model-making

I address questions about values in model-making in engineering, specifically: Might the role of values be attributable solely to interests involved in specifying and using the model? Selected examples illustrate the surprisingly wide variety of things one must take into account in the model-making itself. The notions of system (as used in engineering thermodynamics), and physically similar systems (as used in the physical sciences) are important and powerful in determining what is relevant to an engineering model. Another example (windfarms) illustrates how an idea to completely re-characterize, or reframe, an engineering problem arose during model-making.I employ a qualitative analogue of the notion of physically similar systems. Historical cases can thus be drawn upon; I illustrate with a comparison between a geoengineering proposal to inject, or spray, sulfate aerosols, and two different historical cases involving the spraying of DDT (fire ant eradication; malaria eradication). The current geoengineering proposal is seen to be like the disastrous and counterproductive case, and unlike the successful case, of the spraying of DDT. I conclude by explaining my view that model-making in science is analogous to moral perception in action, drawing on a view in moral theory that has come to be called moral particularism.