Compatible Information as a Natural Information Measure of a Quantum Channel

As a natural quantitative characteristic of mutual information contained in two compatible sets of quantum states considered as input and output, we introduce the Shannon amount of information corresponding to two independent general measurements of all possible quantum states of the input and output. We analyze the physical content of this information measure and its relation to other measures such as the Holevo information and coherent information. In an example of two two-level systems, the most important features of compatible information in the absence and presence of selection for measured states are revealed and discussed.     

Utilization of Information Measure as a Means of Image Thresholding

The entropy method for image thresholding suggested by Kapur et al. has been modified and a more pertinent information measure of the image is obtained. Essentially this consists of viewing the image as a compositum of two fuzzy sets corresponding to the two classes with membership coefficient associated with each gray level a function of its frequency of occurrence as well as its distance from the intermediate threshold selected. An extension of this technique to consider the semantic content of the image is also discussed. The superiority of the suggested method over artificial histograms modelled by Gaussian distributions is demonstrated. Experimental results on several images are also presented to support the validity of the concepts used.     

基于英文概率测度的多表密码分析

传统密码对现代密码学的研究产生了深远而广泛的影响,多表密码作为传统密码的代表,仍然具有许多研究价值.对惟密文攻击,多表密码仍具有一定的安全性,值得进一步的深入研究.采用重合指数等工具,基于英文概率测度,利用英语自然语言的统计特性,对多表密码进行了深入的分析和研究.提出了快速破解多表密码周期的途径和方法,在此基础上,进而提出了快速挖掘密钥向量的若干方法,并就如何合理应用这些方法给出了具体策略.     

Fuzzy Measure and Probability Distributions: Distorted Probabilities

This work studies fuzzy measures and their application to data modeling. We focus on the particular case when fuzzy measures are distorted probabilities. We analyze their properties and introduce a new family of measures (m-dimensional distorted probabilities). The work finishes with the application of two dimensional distorted probabilities to a modeling problem. Results of the application of such fuzzy measure to data modeling using Choquet integral are discussed     

融合后验概率置信度的动态匹配词格检索?

在基于动态匹配词格检索( DMLS)的关键词检测系统中,应用最小编辑距离作为关键词检出的置信度,在提高检出率的同时也增加虚警率。针对此问题,文中提出融合后验概率置信度的动态匹配词格检索方法。该方法首先将基于Lattice的后验概率引入到DMLS的索引建立中,其次应用数据驱动的音素替换、插入和删除代价,实现更灵活的近似匹配,最后通过联合最小编辑距离和后验概率置信度得分进行关键词检测。实验表明,最小编辑距离和后验概率置信度具有一定的互补性,系统的等错误率相对降低。     

信息检索的概率模型

The study of mathematical models on information retrieval is an important area in the Information Retrieval community. Because of the uncertainty characteristic of IR,the probability model based on statistical probability is apromising model from recent to future. Those models can be classified into classical models and probability network models. Several famous models are introduced and their shortcomings are pointed out in this paper. We also clarifythe relationship of these models and introduce a new models based on statistical language model curtly.     

基于信息丢失概率的停车场信息获取算法

随着城市私家车数量的增加,实时获取停车场信息对减少交通拥堵与无效巡游具有重要意义。为此,在含有不同车型的混合网络下,提出一种基于最小信息丢失概率的停车场信息传输算法BMILP,并在该算法的基础上,设计具有停车需求的私家车处于不同道路模式下获取停车场信息的方法。实验结果表明,与BOPBR算法相比,BMILP算法能够降低停车场信息在传输过程中的竞争概率并缩短信息传输时延。     

基于方向信息测度的图像恢复

文章根据人眼对图像平滑区噪声敏感,而对边缘区不敏感这一特性,提出了基于图像信息测度的图像恢复方法。在图像恢复的过程中,在图像的平滑区加大平滑程度,从而减小噪声;而在边缘区,加大图像恢复程度,从而增强边缘,并且考虑了模糊程度随空间变化这一实际。实验证明本文的方法在增大图像边缘恢复程度,减小噪声放大之间能够达到一个很好的折衷,恢复出来的结果图像视觉效果良好。     

一种新的信息系统安全风险概率评估技术研究

研究信息系统安全评估技术,由于信息系统安全问题的日益不确定性和复杂性,传统的仅依靠固定的几个信息系统属性的安全评估方法已不能有效将隐藏在系统中的安全风险监测出来,造成信息系统安全评估的准确性和有效性不高的问题。为了解决上述难题,提出了一种多维属性的信息系统安全评估方法。通过提取信息系统中的多维度属性信息,并采用信息熵和多属性决策算法提取出其中可疑度较高的属性,最后利用筛选出的属性计算出风险概率,完成安全问题的最终评估。仿真结果表明,新方法能够有效提取出隐藏在复杂信息系统中的风险信息,并准确完成对安全风险的评估,取得了满意的结果。     

信息安全策略的原则和方法

制定恰当的安全策略是实现信息系统安全目标的根本保证。本文首先对信息安全策略的原则作了详细的论述,然后以一个“密码安全策略”的示例来阐述安全策略的实现方法。     

监测麻醉深度的脑电信号的近似熵特征研究

近似熵是一种用来量化时间序列复杂度的新方法。实验表明脑电信号的近似熵在监测麻醉深度过程中能很好地反映全麻时的深浅程度，且具有算法简单等特点，特别适用于分析脑电等生物信号。文章主要用近似熵监测了SD大鼠的麻醉过程，结论和事实非常吻合。     

Qualitative Axioms of Uncertainty as a Foundation for Probability and Decision-Making

Although the concept of uncertainty is as old as Epicurus’s writings, and an excellent quantitative theory, with entropy as the measure of uncertainty having been developed in recent times, there has been little exploration of the qualitative theory. The purpose of the present paper is to give a qualitative axiomatization of uncertainty, in the spirit of the many studies of qualitative comparative probability. The qualitative axioms are fundamentally about the uncertainty of a partition of the probability space of events. Of course, it is common to speak of the uncertainty, or randomness, of a random variable, but only the partition defined by the values of the random variable enter into the definition of uncertainty, not the actual values. It is straightforward to add axioms for decision making following the general line of Savage from the 1950s. Indeed, in the spirit of Epicurus, it is really our intuitive feeling about the uncertainty of the future that motivates much of our thinking about decisions. Here, the distinction between the concepts of probability and uncertainty can be made by citing many familiar examples. Without spelling out the technical details, the axiomatization of qualitative probability with uncertainty as the most important primitive concept, it is possible to raise a different kind of question about bounded rationality. This new question is whether or not one should bound the uncertainty in thinking and investigating any detailed framework of decision making. Discussion of this point is certainly different from the question of bounding rationality by not maximizing expected utility. In practice, we naturally bound uncertainty in our analysis of decision-making problems. As in the case of formulating an alternative for maximizing expected utility, so is the case of rational alternatives to maximizing uncertainty. There are several issues to consider. In the spirit of my other work in qualitative probability, I explore alternatives rather than attempt to give a definitive argument for one single solution.     

基于概率粗糙集模型的信息检索

针对经典粗糙集模型难以分类标引空间以及体现类间关联的缺陷,将条件概率关系结合粗糙集理论引入信息检索,提出一种基于概率粗糙集的信息检索模型。定义标引词空间的条件概率关系,自动挖掘概念相似类形成概念空间。定义文档与查询、文档与文档间语义贴近度的计算方法。根据贴近度实现检索匹配结果的排序输出。仿真实例表明了该方法的可行性和有效性。     

Information as a corporate asset

The author explains the need for and necessity of information for running a complex multinational company like Philips. He gives a short overview of the organization and structure of Philips, employing almost 400,000 people, and its world-wide activities. He deals with specific aspects and problems regarding the production, storage and distribution of a large range of Philips products, i.e., how this huge task was tackled in the early EDP days, and what difficulties and troubles had to be overcome. Gradually a new generation of managers has developed who are able to handle the possibilities of computers and automated information systems.However, the multinational Philips covers thousands of separate units which cannot operate on their own. For business-information exchange they have to communicate in a well-defined language based on stadardized data elements. Data-bases, in the future interconnected by a world-wide communications network, contain data that represent valuable and indispensible information for the Philips company: a real corporate asset.With some reserve the writer dares to make predictions about the future of automation, yet mentions some clear trends and draws interesting conclusions.     

规则度量值序列的信息扩散估计研究

对若干连续期间的数据挖掘将形成规则的度量值序列。对度量值序列的参数估计，可获得度量值的基本特征参数，用作评价规则的兴趣度，掌握规则的演化规律。基于信息扩散原理，提出了适用于小样本的度量值扩散估计方法。以度量值序列作为非完备知识样本，讨论在波动和上升(下降)趋势下的序列参数计算。实验表明，该方法准确简便、抗干扰性好。     

模糊集方法在检索评价系统中的应用

评价一个信息检索系统的标准是该系统在多大程度上满足用户的能力。文中从传统的评价信息检索系统的两大标准———计算查全率和查准率的公式出发,结合实际情况,考虑检索结果相关性是一个模糊的概念,为了对这个模糊的概念以客观的度量,应用模糊集的方法对传统的查全率和查准率的公式进行了两种修改。一种修改是在每篇文档对总的查全率和查准率的影响相同时进行的,而第二种修改则是在每一类文档对总的查全率和查准率的影响相同时进行的。这两种修改是对传统的基于二值相关性的检索评价系统的有效扩充。     

信息检索中的带权邻近度度量研究

信息检索需要解决的主要问题是为信息索取者提供相关、准确甚至完整的信息.大量的传统检索模型基于词袋假设进行建模,不考虑查询词之间的相互联系.词项邻近度信息在现有的研究中常被用于提升经典信息检索模型的检索效果,但大部分工作没有考虑查询中各个词重要性的差异.在现代信息检索的查询请求中,查询词之间不仅不完全相互独立,而且分别具有不同的重要程度.因此,在计算邻近度信息时对查询词的重要性进行区分,将有助于提高检索效果.带权邻近度BM25模型(WP-BM25)使用待检索数据集的背景信息对查询词的重要性进行区分,并将带权邻近度度量方法整合到BM25模型中.在TREC评测的3个标准数据集FR88-89,WT2G和WT10G上的一系列对比实验表明,该模型具有较好的鲁棒性,且能够使检索效果得到显著提升.     

Improvement of Pairwise Alignment by an Information Measure

The alignment of amino acid sequences is an important element in the study of life on the level of genome. There exist several methods to align the sequences by computer. In these methods there are several difficulties to judge whether the alignment result is proper. One of such problems is that many results give the same distance, so that often one cannot determine which result should be used. In this paper it is shown that the number of alignment results can be reduced by applying an information measure based on Shannon's mutual entropy.     

浅谈网络信息安全面临的问题和对策

本文主要探讨了网络信息安全中存在的威胁,并提出了相应的应对措施。