基于连接点的3D多角弧匹配

为了满足实际应用的要求，快速地进行多角弧匹配是必要的，因此，提出了一种快速的3D多角弧表示和匹配的方法。该方法是在分析多角弧几何形状基础上，引入连接点的概念，同时，通过在连接点处建立局部直角坐标系来得到每一连接点处的球面坐标。并用连接点的球面坐标集表示多角弧，因这一表示在旋转和平移变换下是不变的，所以可取该球面坐标集作为多角弧匹配的特征集，用该特征集可保持多角弧的几何属性和拓扑结构。这样3D多角弧匹配就降为1D数值串匹配，从而使使匹配变得简单快速。其测量函数为对应连接点间的均方差。实验结果表明，该匹配算法效果良好，并且对于数值污染具有健壮性。     

Influence of display, response, and response set factors upon the storage of spatial information in complex displays.

Immediate recall for spatial information was studied as a function of stimulus load under 2 display formats, 2 response formats, and 2 response set conditions. 4 groups of 10 Ss each served under 15 replications of all response-format, stimulus-load conditions; groups were distinguished on the basis of display format and set. Each S viewed either a spatial or tabular display of 14-26 geometrical stimuli for 16 sec.; he was then required to report, on either a tabular or spatial response form, the location of relevant stimuli. Correct responses and misplacement errors increased more rapidly for the spatial format as more stimuli were presented. Recoding from 1 display to the other response format did not yield serious decrements. Contrary to expectation, response set enhanced all conditions to a nearly equivalent degree. Results are interpreted in terms of the "chunking" hypothesis. (16 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

基于模糊不一致对的多标记属性约简

在实际生活当中,存在着大量的高维多标记数据,为解决维度灾难问题,通常需要约简属性集。针对目前的多标记属性约简算法未考虑标记关系问题,本文提出了一种融合标记关系的模糊不一致对多标记属性约简算法。利用相对熵(KL散度)度量标记之间的关系,定义标记权重,结合标记权重,定义模糊不一致对,考虑到属性对于模糊不一致对的区分性,定义属性重要性并进行属性约简。在8个数据集上的对比实验表明,所提基于模糊不一致对的多标记属性约简算法优于当前的多标记属性约简算法。     

变精度下不完备邻域决策系统的属性约简算法

邻域粗糙集模型在处理完备的数值型数据中得到广泛应用,但针对不完备的数值型和符号型混合数据进行属性约简的讨论相对较少。为此,首先结合邻域粗糙集给出了可变精度模型下不完备邻域决策系统的上、下近似算子及属性约简;然后通过邻域粒化的方法构建了广义邻域下可变精度的粗糙集模型,并提出了一种属性重要度的评价方法;在此基础上,设计出了面向不完备邻域决策系统的属性约简算法,该算法可直接处理不完备的数值型和符号型混合数据;最后,通过实例分析验证了本文提出的算法能够求解出变精度下不完备邻域决策系统的属性约简结果。     

基于粒极值的最简规则提取算法

应用粗糙集的分辨关系,分别从表的行、列2个方面求出每个属性值的分辨、组合能力,以此确定出粒极值,将含粒极值的粒定义为极值粒。应用粒计算理论,以极值粒集为主要运算对象进行粒逻辑运算,使最终的组合粒最简,即所提取的规则最简。实验验证了算法的有效性。     

International and Turkish information technology standardization

The International Organization for Standardization (ISO) is working towards establishing worldwide concensus-oriented standards in information technology. The report reviews the state-of-the-art in information technology standardization. It establishes that the current trend in both hardware and software is towards open platforms and open distributed systems. Networks are evolving towards intelligent networks. Financial institutions and network operating companies are cooperating in ISO committees to prepare security standards. In Turkiye the Turkish Standards Institute (TSE) is active more than ever in promoting the cause of standards. It is recognized that in the information technology area TSE's activities are limited to trying to have the product conformity to the Latin Alphabet No. 5 which includes Turkish characters at all levels. It is the wish of the industry leaders to introduce the present ISO information technology standards and contribute to the future standards. Various mechanisms are discussed on how this can be established.     

The perimeter of a fuzzy set

In pattern recognition one often wants to measure the perimeters of regions in images. This is straightforward if the region is crisply defined, but if it is fuzzy, it is not obvious how its perimeter can be measured. This paper proposes a definition of perimeter for fuzzy subsets of the plane and shows that it reduces to the standard definition if the fuzzy subset is an ordinary subset. The isoperimetric inequality does not generalize to fuzzy subsets, but certain properties of the perimeters of convex sets do generalize to fuzzy perimeters of convex fuzzy subsets.     

Tender Evaluation by Fuzzy Sets

The process of evaluating tenders is considered to be largely dependent on subjective judgment when cost is not the only criterion used. A systematic procedure based on fuzzy set theory and multicriteria modeling is proposed for the selection of bid contracts. The proposed procedure is suitable for a general tender evaluation process that may involve many decision‐making parties and noninteractive multiple criteria. Illustrative examples are given for cases involving three major criteria: cost, present bid information, and past experience of tenderers.     

Extension of efficient predictive control to the nonlinear case

The combined use of the closed‐loop paradigm, an augmented autonomous state space formulation, partial invariance, local affine difference inclusion, and polytopic invariance are deployed in this paper to propose an NMPC algorithm which, unlike earlier algorithms that have to tackle online a nonlinear non‐convex optimization problem, requires the solution of a simple QP. The proposed algorithm is shown to outperform earlier algorithms in respect of size of region of attraction and online computational load. Conversely, for comparable computational loads, the proposed algorithm outperforms earlier algorithms in terms of optimality of dynamic performance. Copyright © 2005 John Wiley & Sons, Ltd.     

On Learning Sets and Functions

This paper presents some results on the probabilistic analysis of learning, illustrating the applicability of these results to settings such as connectionist networks. In particular, it concerns the learning of sets and functions from examples and background information. After a formal statement of the problem, some theorems are provided identifying the conditions necessary and sufficient for efficient learning, with respect to measures of information complexity and computational complexity. Intuitive interpretations of the definitions and theorems are provided.