Reputation management algorithms for DHT-based peer-to-peer environment

This paper addresses the problem of integration of reputation management mechanisms into data look-up and routing processes to cope with some specific security problems in peer-to-peer (P2P) networks based on Distributed Hash Tables (DHT). The goal is to find a mode for “good” peers to cooperate in order to withstand malicious activity of single malevolent peers and their collusions as well.We start from consideration of individual instruments for reputation evaluation that every single peer should be provided with to autonomously define the level of its own trust towards others. We also propose a possible scenario of integration of some existing reputation management techniques and routing mechanisms in a P2P network based on the popular Kademlia algorithm. Then, we present an interaction algorithm that allows peers to obtain the global trust value for each single node through exchanging opinions in the form of local trust values independently calculated by every peer with other members of the community. Such collaboration between peers is indispensable to cope with malicious activity of inconsistent nature specific to DHT-based environment. To render the algorithm resilient to the presence of malicious participants we propose to apply solutions for Byzantine Agreement to exchange opinions among peers. We also provide mathematical evaluations concerning the complexity and computational costs of the proposed algorithm that evidence apparent improvements over the previous one.     

一种伪随机自治服务可生存调度算法

可生存调度算法是服务可生存性研究中的重要内容,在服务器集群系统中引入随机策略,使服务在各个服务器之间的轮转产生不确定效果,可增加攻击者攻击系统的难度,增强系统可生存性。介绍可生存性研究的层次划分,指出可生存调度算法在可生存性研究中的作用,分析现有可生存调度算法的优缺点,提出一种伪随机自治服务可生存调度算法,并分析算法的性能。     

A feature dependent Naive Bayes approach and its application to the software defect prediction problem

Naive Bayes is one of the most widely used algorithms in classification problems because of its simplicity, effectiveness, and robustness. It is suitable for many learning scenarios, such as image classification, fraud detection, web mining, and text classification. Naive Bayes is a probabilistic approach based on assumptions that features are independent of each other and that their weights are equally important. However, in practice, features may be interrelated. In that case, such assumptions may cause a dramatic decrease in performance. In this study, by following preprocessing steps, a Feature Dependent Naive Bayes (FDNB) classification method is proposed. Features are included for calculation as pairs to create dependence between one another. This method was applied to the software defect prediction problem and experiments were carried out using widely recognized NASA PROMISE data sets. The obtained results show that this new method is more successful than the standard Naive Bayes approach and that it has a competitive performance with other feature-weighting techniques. A further aim of this study is to demonstrate that to be reliable, a learning model must be constructed by using only training data, as otherwise misleading results arise from the use of the entire data set.     

支持MongoDB的事务管理方案研究

MongoDB作为一个基于分布式文件存储的数据库,强大的单表查询语言,以及可扩展的高性能数据存储受很用户喜爱,但其没有对事务的完全支持,使得用户对MongoDB的使用处于被动状态.为改善MongoDB对事务管理方面的兼容性,提出一种支持MongoDB事务管理,完善MongoDB功能的方案.利用MQ与守护进程间的消息通信,使守护进程对事务提交或者回滚后的脏数据进行清理,保证了MongoDB在事务管理方面的可用性与安全性.     

Do UML object diagrams affect design comprehensibility? Results from a family of four controlled experiments

Objective: The main objective of our study is to assess whether the use of UML (Unified Modeling Language) object diagrams improves comprehensibility of software design when this kind of diagrams is added to UML class diagrams.Method: We have conducted a family of four controlled experiments. We involved groups of bachelor and master students.Results: Results suggest that the use of object diagrams does not always introduce significant benefits in terms of design comprehensibility. We found that benefits strongly depend on the experience of participants and their familiarity with UML. More experienced participants achieved better design comprehensibility when provided with both class and object diagrams, while less experienced seemed to be damaged when using class and object diagrams together. Results also showed the absence of substantial variations in the time needed to comprehend UML models, with or without object diagrams.Implications: Our results suggest that it is important to be aware and take into account experience and UML familiarity before using object diagrams in software modeling.     

现代汉字形声字声符在普通话中的表音度测查

“形声”作为一种重要的造字方式,构筑了汉字家族中最为庞大的一支。造字之初,形声字以形符表义,以声符表音。随着时代的发展,声符的表音度渐渐发生变化,为人们准确地标音读字造成了一定困难。该文试采用聚类分析的方法,以普通话中3 500常用汉字为对象,结合语言学理论和计算机知识,依据声符表音程度相同、相似和不同制定详细分级标准,并得到每一层级的形声字表和百分数据,从而对现代汉字中形声字声符的表音度情况进行系统、直观而全面地呈现,以期为现代汉字规范的制定和汉语教学提供一定的参考和佐证。     

便携式二氧化碳检测仪的设计

为了探测地震废墟狭小空间内幸存者的呼出气体,设计了便携式二氧化碳检测仪.检测仪采用红外二氧化碳传感器进行检测,通过单片机直接获取数字信号,并具有温度补偿功能.检测结果通过真空荧光显示屏显示,超过预设值触发声光报警.废墟实验表明:该检测仪适用于狭小空间内幸存者的搜索工作.     

面向多核多线程的移动对象连续K近邻查询

针对移动对象的多用户连续K近邻查询处理问题,结合多核多线程技术的发展,提出了一种基于多线程的两阶段多用户连续K近邻查询处理框架.将查询处理分为查询预处理阶段和查询执行阶段,分别执行数据更新任务和查询处理任务,每个阶段都设计了优化cache访问命中率,并利用多线程技术提高多用户连续查询处理并行性的方法及数据结构.提出了一...     

基于扩展的SURF描述符的彩色图像配准技术*

目前大多数图像配准算法都需要先将彩色图像转换为灰度图像再进行图像配准,色彩信息的丢失可能引起图像的误匹配。本文在SURF算法的基础上,提出了构建颜色描述向量扩展SURF描述符,形成ESURF描述符,再进行图像配准的方法。该算法能够充分利用彩色图像的色彩信息,相比大多数算法基于灰度图像的配准方法有更高的鲁棒性,同时继承了SURF算法良好的性能。描述符性能测试和图像配准测试证明：ESURF算法比灰度图像SURF算法在图像尺度变化、旋转、模糊、视角变化、特别是光照变化方面有更高的鲁棒性。