基于任务知识相似性的专家意见综合集成方法

将专家意见的建模过程解释为定性知识的量化,建立了层次化的专家意见模型。提出了基于任务知识相似性的专家意见综合集成方法,从而实现了专家意见的综合集成。     

社团挖掘和话题监控的互动模型研究

社团挖掘是Web信息挖掘领域的重要应用,而话题监控是文本信息研究领域的重要应用,目前这两种技术是各自独立的。为更好地应用于互联网形成的复杂社会网络,将这两种技术结合起来研究,发现了社团和话题之间的关系,创建了社团挖掘和话题监控的静态和动态互动模型,设计了社团挖掘、话题识别以及社团跟踪算法。     

基于Haar小波变换高频特征的图像质量评价算法

在分析人类视觉系统(HVS)特性的基础上,提出一种基于Haar小波变换高频特征的图像质量评价算法(CSSIM)。此方法首先将图像进行小波变换得到不同方向和不同频率的4个频带划分,在3个高频频带比较边缘结构信息上根据人类视觉系统特性,赋予3个频带比较结果相应视觉权重值,最终得到归一化的图像质量评价指标CSSIM。通过非线性回归拟合分析,证明了提出的算法(CSSIM)相对图像结构相似度评价算法(MSSIM)及其改进算法(MGSSIM)更符合人类的主观视觉感受。     

A comprehensive decision making structure for partitioning of make-to-order, make-to-stock and hybrid products

Manufacturing systems produce their products by making use of three main strategies: make-to-order, make-to-stock and hybrid (combination of MTO and MTS strategies) systems. Hence, several products must be considered and partitioned in terms of many different conflicting criteria, leading to a large set of subjective or ambiguous data. Thus, an effective evaluation approach is essential to improve decision quality. In this paper a comprehensive decision, making structure is proposed to choose the proper strategy for producing the products. It consists of a novel hybrid methodology, which combines analytical hierarchy process (AHP) and technique for order performance by similarity to ideal solution (TOPSIS) for partitioning of products. In this study, the rating of each alternative regarding different criteria (throughout the TOPSIS technique) is described by means of linguistic variables. Furthermore, fuzzy AHP methodology is applied to determine the weights of evaluation criteria. At the end, the effectiveness of the proposed model is demonstrated through a case study.     

对动画片的剧本主题研究

本文就动画片剧本主题的发展历史和现状,结合国内外相关领域的成功案例,分析我国动漫产业如何发掘出更多的动画故事,使得我国动画剧本的取材更加多样化,制作出更多的优秀动画作品,从取材致胜,加快我国动画产业的进一步发展。     

话题发现与追踪技术研究

话题发现与追踪以新闻流为处理对象,采用基于事件的信息组织方式进行研究,一直是自然语言处理领域里的热点。该研究借鉴大量相关研究尤其是信息检索中的经典模型和方法,取得了很大成功。首先介绍了话题发现与追踪的主要研究内容、评价方法以及发展历史;然后对其多个研究内容提出一个统一研究框架,并对该框架中的关键技术进行了详细分析;最后指出该领域中的关键问题及难点,并对未来研究做出展望。     

基于OWL的本体集成

提出一种新的本体集成方法。分析了本体集成的原因,阐述了本体集成时应遵循的4条基本原则,并给出了集成的分类,提出了一种基于OWL DL图闭包的本体集成方法。该方法将OWL DL本体抽象为RDFS图模型,根据给定的OWL DL推理规则生成OWL DL本体的图闭包,在此基础上进行本体集成,同时提出了几种计算实体相似度的方法,将本方法与COMA++和FCA-merge进行实验对比,本方法在准确率和召回率上占优势。     

一种基于向量空间模型的主观题批改算法

文字类主观题的自动批改是在线考试系统的一个关键技术。本文在研究主观题自动批改技术现状的基础上,提出了一种基于向量空间模型的短语级句子相似度计算方法,并基于该计算方法设计了一种主观题自动批改算法,经验证,实验结果基本达到预期目标。     

用于图像Hash的视觉相似度客观评价测度

由于评价图像Hash性能时,要求对两幅图像是否在感知上相似做出判断,因此针对这一需求,提出了一种衡量感知相似程度的评价测度。该测度的确定是先对图像进行低通滤波,再进行图像重叠分块;然后运用相关系数检测法计算每一对分块的相似程度,并对相似系数归一化,再分别计算若干个最小和最大的归一化相似系数的乘积;最后用最小相似系数乘积与最大相似系数乘积的比值作为感知相似性的测度。实验结果表明,该测度不仅可有效反映图像视觉质量的变化,而且能较好地区分两幅图像是否存在重要的视觉差异,其对感知相似进行评价的性能优于峰值信噪比。     

Point Set pattern matching ind-dimensions

In this paper we apply computational geometry techniques to obtain an efficient algorithm for the following point set pattern matching problem. Given a setS ofn points and a setP ofk points in thed-dimensional Euclidean space, determine whetherP matches anyk-subset ofS, where a match can be any similarity, i.e., the setP is allowed to undergo translation, rotation, reflection, and global scaling. Motivated by the need to traverse the sets in an orderly fashion to shun exponential complexity, we circumvent the lack of a total order for points in high-dimensional spaces by using an extension of one-dimensional sorting to higher dimensions (which we call circular sorting). This mechanism enables us to achieve the orderly traversal we sought. An optimal algorithm (in time and space) is described for performing circular sorting in arbitrary dimensions. The time complexity of the resulting algorithm for point set pattern matching is O(n logn+kn) for dimension one and O(kn^d) for dimensiond2.Supported in part by CNPq-Conselho Nacional de Desenvolvimento Cientifico e Tecnológico (Brazil) under Grants 200331/79, 300157/90-8, and 500787/91-3.Supported in part by the National Science Foundation under Grant CCR 8901815.