We are promoting the research and development of materials informatics, which realizes the efficiency and sophistication of material research by utilizing a huge amount of experimental data. In order to accelerate new material research and development, it is important to predict the results of new ma-terial development by unimplemented experiments from past experimental data, and for that purpose, a large amount of experimental data generated through various material research processes is used for business. The challenge of material research is to reduce the time for search and acquire data from enormous experiment data. In this paper, we show the experimental da-ta management platform by describing the research and development work-flow with a unified data model. Through prototyping and evaluating the pro-posed technique, we confirmed that the time required for system construction and data acquisition of required experimental data can be reduced by an average of 94.4% in the target work.

This paper introduces the domain of software process improvement (SPI) by including a review of it as an approach that can improve software quality.In addition, it reviews work that has been carried out in the SPI area, as well as problems that can limit how successful SPI initiatives are in various highlighted companies.

文本分类是信息检索的关键问题之一.提取更多的可信反例和构造准确高效的分类器是PU(positive andunlabeled)文本分类的两个重要问题.然而,在现有的可信反例提取方法中,很多方法提取的可信反例数量较少,构建的分类器质量有待提高.分别针对这两个重要步骤提供了一种基于聚类的半监督主动分类方法.与传统的反例提取方法不同,利用聚类技术和正例文档应与反例文档共享尽可能少的特征项这一特点,从未标识数据集中尽可能多地移除正例,从而可以获得更多的可信反例.结合SVM 主动学习和改进的Rocchio 构建分类器,并采用改进的TFIDF(term frequency inverse document frequency)进行特征提取,可以显著提高分类的准确度.分别在3 个不同的数据集中测试了分类结果(RCV1,Reuters-21578,20 Newsgoups).实验结果表明,基于聚类寻找可信反例可以在保持较低错误率的情况下获取更多的可信反例,而且主动学习方法的引入也显著提升了分类精度.

在传统的公钥加密方案中,一旦解密密钥泄漏,系统的安全性将完全丧失.特别是随着越来越多的加密系统被应用到移动的、安全性低的设备中,密钥泄漏显得难以避免.入侵容忍公钥加密的提出就是为了减小密钥泄漏对加密系统的危害,具有比前向安全加密、密钥隔离加密更强的安全性.在这种体制下,整个生命周期被分割成离散的时间阶段,公钥固定不变,密钥信息分享在解密者和基地中,前者独立完成解密操作,而后者则在每个时间周期中提供一个更新信息来帮助演化解密密钥.此外,每个时间段内有多次密钥刷新的操作,可以刷新解密者的密钥和基密钥.当解密者和基地被入侵时,只要不是同时被入侵,安全性就可以得到保证.即使入侵者同时入侵解密者和基地,也不会影响以前时间段密文的安全性.提出了一个入侵容忍公钥加密方案,所有费用参数关于总共时间段数的复杂性均不超过对数的平方.证明了该方案是标准模型下安全的.这是一个不需要随机预言的可证明安全的入侵容忍公钥加密方案.;In traditional public-key encryption schemes, security guarantees will be fully lost once decryption secret keys are exposed. With the ever-increase in encryption systems used in mobile and low secuirity devices, key exposure seems unavoidable. An intrusionresilient public-key encryption is proposed to mitigate the damage for the encryption systems brought by key exposure, which provides more security than the forward-secure encryption and key-insulated encryption. In its primitive, the whole lifetime is divided into discrete periods where the public key is fixed. Secret keys are shared in a decrypter and a base. The former performs the decrypting operations on his own while the latter provides an updated message to help evolve secret keys in each period. Furthermore, multiple operations of refresh secret keys are performed to refresh decrypter and base secrets periodically. The security can be preserved when both the user and base are compromised, as long as they are not compromised simultaneously. In addition, the simultaneous compromise doesn't affect the security of the ciphertext generated in previous periods. This paper proposes an intrusion-resilient public-key encryption scheme. All the parameters in this scheme have at most a log-squared complexity in terms of the total number of time periods. The proposed scheme is proven to be secure in the standard model and is a provably secure intrusion-resilient public-key encryption scheme without random oracles.

以僵尸网络为载体的各种网络攻击活动是目前互联网所面临的最为严重的安全威胁之一.虽然近年来这方面的研究取得了显著的进展,但是由于僵尸网络不断演化、越来越复杂和隐蔽以及网络和系统体系结构的限制给检测和防御带来的困难,如何有效应对僵尸网络的威胁仍是一项持续而具有挑战性的课题.首先从僵尸网络的传播、攻击以及命令与控制这3 个方面介绍了近年来僵尸网络工作机制的发展,然后从监测、工作机制分析、特征分析、检测和主动遏制这5 个环节对僵尸网络防御方面的研究进行总结和分析,并对目前的防御方法的局限、僵尸网络的发展趋势和进一步的研究方向进行了讨论.;Botnets are one of the most serious threats to the Internet. Researchers have done plenty of research and made significant progress. However, botnets keep evolving and have become more and more sophisticated. Due to the underlying security limitation of current system and Internet architecture, and the complexity of botnet itself, how to effectively counter the global threat of botnets is still a very challenging issue. This paper first introduces the evolving of botnet’s propagation, attack, command, and control mechanisms. Then the paper summarizes recent advances of botnet defense research and categorizes into five areas: Botnet monitoring, botnet infiltration, analysis of botnet characteristics, botnet detection and botnet disruption. The limitation of current botnet defense techniques, the evolving trend of botnet, and some possible directions for future research are also discussed.

利用最大团问题解空间特殊的结构特征,提出一种基于染色划分构建高维约束指导局部搜索移动方向的改进RLS 算法——RLS-II 算法,该算法提高了局部搜索向最优解靠近的概率.基于吸收态Markov 链理论,建立了RLS 和RLS-II 算法求解最大团问题的数学模型,分析了两种算法的吸收时间,并在77 个标准测试算例上对分析结果进行了实验验证.理论分析及实验结果都表明,染色划分过滤确实能够有效改善RLS 算法的性能,且平均染色组长度越大,性能改进的概率和幅度就越大.

移动模型是Ad Hoc 网络区别于其他形式网络的重要标志,对其产生的动态网络特性(简称动态特性)进行评估,是研究Ad Hoc 网络的协议仿真和网络相关技术(如拓扑控制和网络性能测量等)的基础性问题.在已有研究的基础上,改进了网络的模型化描述,克服了以往模型无法很好地描述相关联的时空动态特性的缺陷,并在此基础上,提出了移动模型通用的可量化时空动态特性评估方法.通过构建节点空间位置分布,建立网络拓扑时空动态特性的分析模型,深入研究了几种移动模型的动态性.提出一种圆周曲线移动模型,弥补了以往移动模型难以描述现实的曲线移动场景.仿真实验结果表明,该方法能够有效地对现有移动模型的动态性进行评估.实验结果表明,圆周曲线移动模型与以往移动模型相比,具有良好的时空动态特性.;Mobility model is a crucial feature to differentiate ad hoc networks from other networks. The evaluation of dynamic network properties caused by mobility model is the basic problem for research on protocol simulation and related technologies, such as topology control and network performance measurement, in ad hoc networks. In this paper, a model-based description of ad hoc networks is improved to avoid the drawbacks in describing the corresponding space-time dynamic properties. Then a universal evaluation method with measureable space-time dynamic properties is presented. The dynamic properties of several mobility models are studied by constructing the space distribution of the nodes and designing the models for analyzing the topological space-time dynamic properties. Furthermore, a novel mobility model based on circular curve movement is proposed to improve the flexibility and the capability to describe realistic scenarios with curve movement. Simulation with NS2 shows the presented method can effectively evaluate the characteristics of the existing mobility models. The numerical results also show that the proposed mobility model has good dynamic properties in comparison with the existing mobility models.

回答集编程(answer set programming,ASP)是一种回答集语义下的逻辑编程范例,可应用于非单调推理,叙述式问题求解等领域.本文为ASP提出并实现了一种破圈启发方法与一种基部限制式前向搜索过程,所得到的系统称为LPS.实验结果显示,相对于其他经典的ASP系统,LPS能够有效地解决处于相变难区域中的逻辑程序,通常这些程序被认为是计算困难的.除此以外,通过使用被称为动态变元过滤(dynamic variable filtering,DVF)的技术,LPS可以在计算过程中极大地缩小搜索树的尺寸.

This paper proposes a resource-aware performance diagnostic method. For transactions in Web applications, the proposed method constructs performance profile chains based on the resource service time, which is stable for different workload characteristics. According to the anomaly of resource service time at application runtime, the proposed method provides an efficient solution to performance anomaly detection, location and diagnosis. Experimental results show that this method can effectively detect performance anomalies caused by different resource bottlenecks with changing workload characteristics.

从3个方面改进了不确定规划(non-deterministic planning,简称NDP)中的观测约简:一是如何找最小观测集合(minimal observation set,简称MOS),二是如何在观测代价不均等时找最优观测集合(optimal observation set,简称OOS),三是如何找到容错的OOS.通过MOS问题和图论中的最小覆盖集问题(minimal set cover,简称MSC)的类似性,可证MOS是NP难的问题,还可参考MSC算法得出时间复杂性不超过O(2^mm²)且不低于Ω(2^m?1)的算法,其中m是观测的个数.通过使用整数规划(integer programming,简称IP)技术,可找到OOS以及容错的OOS.可以证明,上述算法能够保证找到解,并且能够保证解的最优性.