随机产生式系统的启发式图搜索算法RA及A的推广

本文首先提出了随机产生式系统RPS的概念,然后,运用三角模的概念,得到了随机产生式系统RPS的启发式图搜索算法RA及推广的A算法,并对RA及推广的A的可采纳性给出了证明.     

基于概念网络的文本分类研究

提出了一种基于概念网络和主题概念树的文本分类算法。该算法可以根据关联度传播模型对未知文本中的一些概念进行一定程度上的语义复合。     

一种基于概念信息量的相似度传播算法

相似度传播在本体概念相似度计算中有着非常重要的作用.然而,目前常见的相似度传播算法大都采用了固定比例的相似度传播值,没有对相似度转播值进行合理的定量分析.针对此问题,提出了基于概念信息量的相似度传播算法,该算法根据匹配节点的概念信息量大小来判断其子父节点匹配概率大小,通过匹配概率大小调整相似度传播值,从而进行更精确的相似度传播.理论分析与实验结果证明了该算法是有效的.     

基于超像素的自然图像抠图

现有的自然图像抠图算法可以分为三类:基于采样、基于传播和基于机器学习。通常为传播算法设计一个有效的像素特征非常困难,也一直是影响算法结果好坏的重要因素。本文探索将超像素作为特征应用在传播算法上效果,并设计了新的函数来衡量两个超像素之间的相似性。实验结果表明本文提出的方法能更有效地区分前背景像素,建立更准确的全局像素关系,并在标准测试集上取得了领先效果。     

基于资源分配指标的最大约束社区发现算法

在复杂网络中的社区发现一直受到广泛的关注,基于模块度最大化的方法是目前流行的社区发现技术。提出一种基于资源分配(RA)指标和多步贪婪凝聚策略的模块度最大化社区发现算法RALPA(Resource Allocation-based of Label propagation Algorithm)。该算法利用准确衡量节点间相似性的RA指标,通过最大约束标记传播模型使社区内部节点拥有较高的相似性,与社区外部的节点拥有较低的相似性。然后,通过多步贪婪凝聚策略将划分模块度增加最大的多对小社区进行合并。实验结果表明,该算法不仅避免了对节点更新顺序的敏感和易得到平凡解的问题,而且提高了算法的稳定性和社区划分的精度。     

基于事件树和PRA的民航机场恐怖袭击风险评估模型

为了降低民航机场恐怖袭击事件发生的概率和风险,提出了一种基于事件树和PRA的民航机场恐怖袭击风险评估模型,利用该模型对恐怖袭击方式、入侵路径、损失概率和遇袭风险等事件进行仿真,通过GTD和其他开源数据库,结合Delphi咨询法,以定量分析的方法,对某民航机场潜在的各类恐袭事件损失概率和遇袭风险进行了评估。仿真结果表明,小型无人机很可能被恐怖分子用于未来的恐袭活动中,应加强机场周界防御和监测措施。     

一种新型基于用户指导的多关系关联规则挖掘算法

本文提出了一种基于用户指导的多关系关联规则挖掘算法,借鉴有向图的概念动态的选择最优关键表,并利用元组ID传播的思想使多表间无需物理连接而能直接进行关联规则挖掘,同时引入用户指导的概念,提高了用户的满意程度及海量数据挖掘的效率和精确度,该算法能够直接支持关系数据库,且运行效率远远高于基于ILP技术的多关系关联规则挖掘算法,     

基于PRA方法风险评价系统的设计与研究*

针对动态系统安全性分析的需要,设计并实现了一个新的风险评价系统.该系统在PRA方法的基础上,在风险评估过程中综合运用多种动态系统的安全性建模和危险分析技术,能够较好地适应动态系统风险评估的需求.     

本体映射中结构策略改进算法

相似度传播是本体映射中结构策略的实质内容。然而,目前的相似度传播算法却没有对相似度传播值进行合理的定量分析。针对此问题,提出了基于概念信息量的相似度传播算法,该算法根据已匹配节点的概念信息量大小来判断其子父节点匹配概率大小,通过匹配概率大小调整相似度传播值,从而进行更精确的相似度传播,提高结构策略的可信度。理论分析与实验结果证明了该算法是有效的。     

基于向量表示和标签传播的半监督短文本数据流分类算法

社交网络平台产生海量的短文本数据流,具有快速、海量、概念漂移、文本长度短小、类标签大量缺失等特点.为此,文中提出基于向量表示和标签传播的半监督短文本数据流分类算法,可对仅含少量有标记数据的数据集进行有效分类.同时,为了适应概念漂移,提出基于聚类簇的概念漂移检测算法.在实际短文本数据流上的实验表明,相比半监督分类算法和半监督数据流分类算法,文中算法不仅提高分类精度和宏平均,还能快速适应数据流中的概念漂移.     

PRA*: Massively Parallel Heuristic Search

In this paper we describe a variant of A* search designed to run on the massively parallel, SIMD Connection Machine (CM-2). The algorithm is designed to run in a limited memory by the use of a retraction technique which allows nodes with poor heuristic values to be removed from the open list until such time as they may need reexpansion, more promising paths having failed. Our algorithm, called PRA* (for Parallel Retraction A*), is designed to maximize use of the Connection Machine′s memory and processors. In addition, the algorithm is guaranteed to return an optimal path when an admissible heuristic is used. Results comparing PRA* to Korf′s IDA* for the fifteen puzzle show significantly fewer node expansions for PRA*. In addition, empirical results show significant parallel speedups, indicative of the algorithm′s design for high processor utilization.     

基于PRA的网络安全风险评估模型

概率风险分析被广泛应用于社会各领域,如交通、能源、化工处理、航天、军事等。文章采用概率风险分析的方法,对网络的逻辑构成、网络攻击和攻击结果进行分析,通过故障树描述了网络系统被攻击的原因与途径,并建立了风险评估模型。     

Optimizing ontology alignment through Memetic Algorithm based on Partial Reference Alignment

All the state of the art approaches based on evolutionary algorithm (EA) for addressing the meta-matching problem in ontology alignment require the domain expert to provide a reference alignment (RA) between two ontologies in advance. Since the RA is very expensive to obtain especially when the scale of ontology is very large, in this paper, we propose to use the Partial Reference Alignment (PRA) built by clustering-based approach to take the place of RA in the process of using evolutionary approach. Then a problem-specific Memetic Algorithm (MA) is proposed to address the meta-matching problem by optimizing the aggregation of three different basic similarity measures (Syntactic Measure, Linguistic Measure and Taxonomy based Measure) into a single similarity metric. The experimental results have shown that using PRA constructed by our approach in most cases leads to higher quality of solution than using PRA built in randomly selecting classes from ontology and the quality of solution is very close to the approach using RA where the precision value of solution is generally high. Comparing to the state of the art ontology matching systems, our approach is able to obtain more accurate results. Moreover, our approach’s performance is better than GOAL approach based on Genetic Algorithm (GA) and RA with the average improvement up to 50.61%. Therefore, the proposed approach is both effective.     

启发式图搜索算法RA~*的改进算法IRA~*及IRA′

本文在文[1]基础上,对两种RA~*算法进行了比较研究,通过引入感兴趣集,给出了RA~*算法的改进算法IRA~*和IRA’,并且证明了IRA~*算法的可采纳性.从算法所扩展的结点数目这一角度来看,IRA~*算法明显优于RA~*算法.若感兴趣集并不包含最佳路径上的结点,则IRA~*算法的变形——IRA’算法可用来寻找一条较佳的求解路径.     

应用于机器人路径规划的双向时效A*算法

针对时效A*算法为了大幅减少算法时间,导致路径规划长度增加和路径锯齿过多的问题,提出一种改进的双向时效A*算法,该方法将从起点和终点同时运行时效A*算法寻找路径,并采用多近邻栅格距离计算方案;同时,根据不同环境地图对传统A*算法、时效A*算法和双向时效A*算法运行结果进行对比研究及分析;最后,制定算法时间、路径长度两个指标来评判算法的优劣。实验结果显示,双向时效A*算法相对于传统A*算法,算法时间最大减少76.8%,相对于时效A*算法,时间最大减少55.4%,并解决了时效A*算法规划路径距离增加、路径不够平滑的问题。     

双向启发式图搜索算法BRA^＊之研究

本文在[1]中基于模运算,提出了随机产生式系统的启发式图搜索算法RA~*.本文提出一个随机产生式系统的双向搜索的启发式图搜索算法BRA~*,证明了算法BRA~*的可采纳性,并得到了一些新的可采纳性结果.算法BRA~*的搜索效率比算法RA~*高.若启发式估价函数满足单调性限制,通过使用NP操作,则算法BRA~*的搜索空间将进一步减少.     

论山水画美学

绘画的哲学层次,道与天地的概念,贯穿在整个山水画美学之中。我们知道水的灵动,山的厚重,山水相联,又相互映衬而构成一幅幅出色的山水画。中国的山水画强调山与水的交融,凸显和谐之美。本文通过对山水树木,云天湖泊与时间和空间的特殊画面结构将大自然的这三个世界联结在一起的,提出山水画美学观。     

Towards a Trajectory Planning Concept: Augmenting Path Planning Methods by Considering Speed Limit Constraints

Trajectory planning is an essential part of systems controlling autonomous entities such as vehicles or robots. It requires not only finding spatial curves but also that dynamic properties of the vehicles (such as speed limits for certain maneuvers) must be followed. In this paper, we present an approach for augmenting existing path planning methods to support basic dynamic constraints, concretely speed limit constraints. We apply this approach to the well known A* and state-of-the-art Theta* and Lazy Theta* path planning algorithms. We use a concept of trajectory planning based on a modular architecture in which spatial and dynamic parts can be easily implemented. This concept allows dynamic aspects to be processed during planning. Existing systems based on a similar concept usually add dynamics (velocity) into spatial curves in a post-processing step which might be inappropriate when the curves do not follow the dynamics. Many existing trajectory planning approaches, especially in mobile robotics, encode dynamic aspects directly in the representation (e.g. in the form of regular lattices) which requires a precise knowledge of the environmental and dynamic properties of particular autonomous entities making designing and implementing such trajectory planning approaches quite difficult. The concept of trajectory planning we implemented might not be as precise but the modular architecture makes the design and implementation easier because we can use (modified) well known path planning methods and define models of dynamics of autonomous entities separately. This seems to be appropriate for simulations used in feasibility studies for some complex autonomous systems or in computer games etc. Our basic implementation of the augmented A*, Theta* and Lazy Theta* algorithms is also experimentally evaluated. We compare (i) the augmented and basic A*, Theta* and Lazy Theta* algorithms and (ii) optimizing of augmented Theta* and Lazy Theta* for distance (the trajectory length) and duration (time needed to move through the trajectory).     

Scratch my back and I'll scratch yours: The impact of user effort and recommendation agent effort on perceived recommendation agent quality

Product recommendation agents (RAs) assist online firms to adapt their suggested offers to users’ preferences, thereby lowering users’ decision effort. The concept of effort is central in decision-making, yet it remains unclear whether it should be regarded as a cost or as a benefit improving the odds of a better decision. Building on Social Exchange theory, we suggest that interactions between users and RAs are driven by the concepts of perceived interdependence and reciprocity. We suggest that perceived user effort decreases the perceived RA quality, whereas RA effort increases the perceived RA quality. We conducted two experimental studies across different contexts. We found that users evaluate RAs based on their own expended effort, in relation to how much effort they perceived the RA has put into the process of generating recommendations. Such an effect is attenuated by users’ familiarity with the product context. Our findings offer important insights into how online firms can improve the use of their RAs.