异常点挖掘研究进展

异常点是数据集中与其他数据显著不同的数据.一个人的噪声对另一个人而言可能是有用的数据,因此,随着人们对数据质量、欺诈检测、网络入侵、故障诊断、自动军事侦察等问题的关注, 异常点挖掘在信息科学研究领域日益受到重视.在充分调研国内外异常点挖掘研究文献基础上,系统地综述了数据库研究领域中异常点挖掘的研究现状,对已有各种异常点挖掘方法进行了总结和比较,并结合当前研究热点,展望了异常点挖掘未来的研究方向及其面临的挑战.     

基于弯曲测试的纳结构机械力学特性的表征

利用原子力显微镜对聚焦离子束刻蚀技术制备的不同纵横比的氮化硅纳米梁结构进行了弯曲测试,并讨论了一种标定原子力显微镜微悬臂梁弹簧常数的方法.实验中严格界定了欧拉细长梁的条件,在分析了纳米梁组纵横比对测试结果影响的基础上,讨论了如何修正弯曲测试结果.弯曲测试得到了纳米梁的平面模数,并进一步推算氮化硅材料的杨氏模量.测试结果验证了基于原子力显微镜系统的弯曲方法测试纳米薄膜材料杨氏模量的可行性.     

一种基于元启发式策略的迭代自学习K-Means算法

类内误差平方和最小化的聚类准则求解是NP难问题,K-Means采用的迭代重定位方法本质上是一种局部搜索的爬山算法,因此聚类结果对初始代表点的选择非常敏感,只能保证局部最优.为此,引入元启发式策略,通过建立评估函数对K-Means初始代表点和目标函数之间的依赖关系进行近似,然后利用近似评估函数指导新的初始代表点的选择,构成一种迭代自学习框架下的K-Means算法.实验表明算法可以很好地克服K-Means对初始代表点的依赖性,获得较高质量的聚类结果.     

Mechanical Geometry Theorem Proving Based on Groebner Bases

A new method for the mechanical elementary geometry theorem proving is presented by using Groebner bases of polynomial ideals.It has two main advantages over the approach proposed in literature:(i)It is complete and not a refutational procdure;(ii) The subcases of the geometry statements which are not generally true can be differentiated clearly.     

Towards the Semantics and Verification of BPEL4WS

In this paper, we discuss the semantics of BPEL4WS language which is a de facto standard for specifying and execution workflow specification for web service composition and orchestration. We propose a language μ-BPEL that includes most primitive and structured activities of BPEL4WS, and define its semantics. As the Timed Automata (TA) is powerful in designing real-time models with multiple clocks and has well developed automatic tool support, we define a map from μ-BPEL into composable TA. Therefore, the properties we want to check can be verified in TA network correspondingly. Furthermore, we prove that the mapping from μ-BPEL to TA is a simulation, which means that the TA network simulates correctly the corresponding μ-BPEL specification. The case study with model checker Uppaal shows that our method is effective, and a Java supporting tool based on Uppaal model checker engine has been developed.     

Web video thumbnail recommendation with content-aware analysis and query-sensitive matching

In this paper, a unified and adaptive web video thumbnail recommendation framework is proposed, which recommends thumbnails both for video owners and browsers on the basis of image quality assessment, image accessibility analysis, video content representativeness analysis and query-sensitive matching. At the very start, video shot detection is performed and the highest image quality video frame is extracted as the key frame for each shot on the basis of our proposed image quality assessment method. These key frames are utilized as the thumbnail candidates for the following processes. In the image quality assessment, the normalized variance autofocusing function is employed to evaluate the image blur and ensures that the selected video thumbnail candidates are clear and have high image quality. For accessibility analysis, color moment, visual salience and texture are used with a support vector regression model to predict the candidates’ accessibility score, which ensures that the recommended thumbnail’s ROIs are big enough and it is very accessible for users. For content representativeness analysis, the mutual reinforcement algorithm is adopted in the entire video to obtain the candidates’ representativeness score, which ensures that the final thumbnail is representative enough for users to catch the main video contents at a glance. Considering browsers’ query intent, a relevant model is designed to recommend more personalized thumbnails for certain browsers. Finally, by flexibly fusing the above analysis results, the final adaptive recommendation work is accomplished. Experimental results and subjective evaluations demonstrate the effectiveness of the proposed approach. Compared with the existing web video thumbnail generation methods, the thumbnails for video owners not only reflect the contents of the video better, but also make users feel more comfortable. The thumbnails for video browsers directly reflect their preference, which greatly enhances their user experience.     

特征模型定制的一种自动传播策略与选择性回退机制

面向特征的分析方法已被主流的软件复用方法接受.其基本思想是将软件的特征作为软件解决需求问题的基本要素,使用特征以及特征之间的关系(这两者的组合也被称为特征模型)来组织软件的需求.特征模型良好的可剪裁性和可扩展性,为领域需求的复用提供了有力的支持.在当前的面向特征的方法中,特征模型上的一个定制决策会影响其他相关特征的绑定状态.如何使用特征模型之间内在的关联关系来实现特征模型的定制的自动传播算法与选择性回退算法是一个值得进行深入研究的课题.作者给出了特征模型定制的一种自动传播策略与选择性回退机制,针对特征模型中存在的二元约束与组合约束进行分析,介绍该算法的实现细节.     

一种社交网络虚假信息传播控制方法

针对社交网络中日渐泛滥的虚假信息,提出了一种社交网络虚假信息传播控制方法Fidic,针对社交网络中虚假信息传播时途经的用户序列,该方法基于PageRank的思想并结合用户传播虚假信息时的指向关系来对用户进行评级,用户的评级越高表明其在虚假信息传播过程中起到的作用越大.该方法通过控制高评级用户的传播行为从而缩小虚假信息传播的覆盖面,最终达到遏制甚至消除虚假信息传播的目的.实验证明该方法既可以获得较高的精度,而且方便可行,具有较高实用价值.     

基于机载多普勒雷达数据的飓风内部三维风场反演算法

利用2002年NOAA海洋飓风实验中三维风雨微波成像仪(IWRAP,UMass Imaging Wind and Rain Airborne Profiler)数据,提出了两种自适应卡尔曼滤波算法进行飓风三维风矢量的反演。为验证不同自适应滤波算法的准确性,通过对实际风场以及相应的机载雷达观测数据进行仿真,得到一种最优自适应滤波;文中对滤波结果误差进行了初步分析,探讨了误差来源。根据仿真结果,利用最优自适应滤波对2002年10月2日飓风莉莉多普勒雷达测量数据进行处理,滤波得到的三维风场与本次实验中GPS下投式探空仪的三维风场同步测量数据,多波段微波辐射计的海面风场测量数据,以及飞机高度风场数据进行了比较。