利用收益预测与策略梯度两阶段众包评论集成

近年来随着互联网的飞速发展,人们频繁地在网络上发布关于某一特定对象的评论内容,快速掌握众包评论文本的关键信息对决策制定、服务调整有着重要作用,对众包评论文本集成进行深入研究亦显得十分必要.众包评论文本集成旨在将不同评论者对同一对象的评论内容以既定压缩率整合成较短的集成文本,从而根据大众认知形成关于特定对象较为匹配的内容描述.针对该问题提出了一种利用收益预测与策略梯度的两阶段众包评论集成方法.该方法不依赖于任何人工真值,仅提供源众包评论文档,由代理根据收益经验自行抽取关键语句形成众包评论集成文档.具体而言,第一阶段以语句相关性与冗余性衡量集成文档内容质量,以此作为收益,利用Q-值学习预测出从当前语句选择起直至评论集成结束时所产生的长期收益,由此指导代理学习最优语句选择策略;在此基础上,第二阶段以集成文档情感强度为收益,利用策略梯度(上升)进一步调整第一阶段代理习得的语句选择策略,使得代理所产生集成文本在具备一定内容质量同时,从客观角度突显文本情感强度,更明确反映出评论者所持有的情感态度.实验结果表明,与现有相关方法相比,所提出方法在评论文本集成内容质量与情感强度方面总体取得最优,且产生集成文档所耗费时长仍控制在可接受范围之内.     

基于ASP.NET心理测量系统的设计与实现

为研究人的心理活动,可以采集心理测量数据,运用测试法进行研究。阐述了心理测量的必要性及ASP.NET技术优势,着重介绍了系统结构设计和功能设计,并实现了基于ASP.NET心理测量系统。     

L2(Rd)(d≥3)空间非张量积样条小波基的构造

本文提出了 L2 (Rd)空间小波基的一种新型构造方法 ,此方法构造出来的小波基不能由张量积产生。     

Implementing fault-tolerance via modular redundancy with comparison

An NMRC (N-modular redundancy with comparison) system is presented. It surpasses all existing NMR systems in fault tolerance capability. The extra logic needed by NMRC is simpler than that of the other NMR systems. The relation between the interconnection topology and the fault tolerance capability of the NMRC system investigated. Three types of optimal NMRC systems and their characterization and structure are studied. NMRC can be viewed as a diagnosable system. The comparison approach is applied to t₁/s-diagnosable systems, whereas previously it had been applied only to t₀ - and t₁-diagnosable systems. A laboratory 3-MRC system has been built as a node computer for a fault-tolerant multicomputer system for industrial process control. The test results confirm the high reliability and effectiveness of NMRC     

A Reliable and Fault-Tolerant Interconnection Network

An interconnection network with multistage redundant paths is introduced for using in high-performance multiprocessor systems.The routing algorithm of the proposed network is simple and dynamically reroutable.The analysis of the fault-tolerance and performance of the network are given.It is shown that the probability of acceptance and the performance-to-cost ration of the network are better than those of F and Gamma Networks.Another advantages of the proposed network is the smaller amount of interstage links compared with F network.     

Access Control Policy Based on Friend Circle

Nowadays, the scale of the user’s personal social network (personal network, a network of the user and their friends, where the user we call “center user”) is becoming larger and more complex. It is difficult to find a suitable way to manage them automatically. In order to solve this problem, we propose an access control model for social network to protect the privacy of the central users, which achieves the access control accurately and automatically. Based on the hybrid friend circle detection algorithm, we consider the aspects of direct judgment, indirect trust judgment and malicious users, a set of multi-angle control method which could be adapted to the social network environment is proposed. Finally, we propose the solution to the possible conflict of rights in the right control, and assign the rights reasonably in the case of guaranteeing the privacy of the users.     

面向连贯性强化的无真值依赖文本摘要模型

自动文本摘要技术旨在凝练给定文本,以篇幅较短的摘要有效反映出原文核心内容.现阶段,生成型文本摘要技术因能够以更加灵活丰富的词汇对原文进行转述,已成为文本摘要领域的研究热点.然而,现有生成型文本摘要模型在产生摘要语句时涉及对原有词汇的重组与新词的添加,易造成摘要语句不连贯、可读性低.此外,通过传统基于已标注数据的有监督训...     

三元比较模型中若干问题的研究

随着网络以及多处理机系统应用的日益普及,这些系统的维护问题也越来越突出。系统故障诊断为系统的维护提供了重要的理论基础,它的基本思想是:首先让系统中的处理机互相测,然后对测试结果进行分析,找出故障处理机。     

Novel mislabeled training data detection algorithm

As a kind of noise, mislabeled training data exist in many applications. Because of their negative effects on learning, many filter techniques have been proposed to identify and eliminate them. Ensemble learning-based filter (EnFilter) is the most widely used filter which employs ensemble classifiers. In EnFilter, first the noisy training dataset is divided into several subsets. Each noisy subset is then checked by the multiple classifiers which are trained based on other noisy subsets. It is noted that since the training data used to train multiple classifiers are noisy, the quality of these classifiers cannot be guaranteed, which might generate poor noise identification result. This problem is more serious when the noise ratio in the training dataset is high. To solve this problem, a straightforward but effective approach is proposed in this work. Instead of using noisy data to train the classifiers, nearly noise-free (NNF) data are used since they are supposed to train more reliable classifiers. To this end, a novel NNF data extraction approach is also proposed. Experimental results on a set of benchmark datasets illustrate the utility of our proposed approach.