共查询到20条相似文献,搜索用时 109 毫秒
1.
为发现语义Web使用记录中所蕴含的有效信息,本文提出了一种挖掘日志本体频繁Web访问模式的方法。该方法引入应用访问规则集和观察集分别表示日志信息动态变化的语义规则和使用事实,并在DL安全的限定下将日志本体和应用访问规则集相结合构成一个推理过程可判定的混合知识库。在此基础上,利用日志本体中事件整分关系的语义构建访问模式学习的事务模型,并采用ILP的方法学习生成频繁用户访问模式树,解决了推理访问模式中非描述逻辑原子的问题。实验结果表明该方法的可用性和有效性。 相似文献
2.
基于兴趣度的Web用户聚类方法 总被引:1,自引:1,他引:0
现有的Web用户聚类方法都是通过对用户喜好页面的访问模式分析来建立用户聚类,没有充分考虑时间意识、用户兴趣、用户访问模式之间的关系与影响.针对这一问题,在时间意识的Web用户聚类基础之上,提出了基于兴趣度的Web用户聚类方法.通过对日志文件中的用户访问模式进行分析,计算用户兴趣度.结合渐进遗忘算法,对用户兴趣爱好进行调整与更新,并在此基础上对用户进行聚类.实验表明,本方法能够更好地分析用户访问模式,更准确地计算用户兴趣,具有更好的聚类效果. 相似文献
3.
目前Web服务架构中用于Web服务发现的UDDI(统一描述、发现和集成)技术主要是基于关键词查找的自身缺陷,UDDI技术不支持概念问的推理,这给用户的服务查找带来不便.因此作者在分析基于UDDI 的Web服务发现机制及其缺陷的基础上,着重研究了基于语义的Web服务发现系统的体系结构,通过引入Web服务质量本体,提出一种改进的语义Web服务发现方法. 相似文献
4.
5.
6.
7.
《现代电子技术》2016,(16):14-18
在Web网络环境下,传统信息检索方法仅依据简单的字和词进行匹配,未考虑知识的描述、处理以及理解等性能,检索质量和效率低。因此,设计了基于Web的语义检索平台,其由数据层、数据访问层、业务逻辑层、控制层和人机接口层组成。业务逻辑层依据数据访问层操作数据,并将数据反馈给控制层;控制层是用户申请和业务逻辑操作间的调控器;人机接口层是用户同检索平台间实现交互的桥梁,用户通过该层中的操作界面完成信息的检索。分析Jena在语义检索平台中的作用,并在软件设计部分,分析通过Jena实现语义检索平台数据检索的过程和其中的关键代码。实验结果表明所设计的语义检索平台具有较高的检索质量和效率。 相似文献
8.
一种基于QoS的语义Web服务本体模型 总被引:1,自引:0,他引:1
语义Web服务是结合语义Web和Web服务两方面技术的一门新的研究课题.文中利用语义网的本体技术,在综合考虑为用户选择提供最佳服务和添加语义描述的基础上,分析设计了一种基于Oos的语义Web服务本体模型,并针对票务服务领域进行了实例设计和语义描述. 相似文献
9.
10.
国伟 《信息技术与信息化》2012,(6)
一个网站有多个网页组成,网站上的信息分布在这些网页上,不同的用户对不同的网页上的信息感兴趣,如何满足用户对敏感网页的高效访问,我们提出一个基于使用挖掘的Web站点个性化信息服务系统来解决问题.利用Web使用挖掘技术来分析用户的浏览模式,根据用户的当前访问需要,自动实时地为用户提供推荐页面,实现个性化服务. 相似文献
11.
12.
13.
A Web site depends on more than its ability to attract. Crucial to its success is understanding what Web site users need and want, and consequently choosing appropriate content, structure, and navigational support. In creating Web based information, professional communicators are able to incorporate the kinds of textual and structural support for readers that is impossible to achieve in hardcopy. This reader-centered approach is familiar to technical communicators, but what about readers who have special needs? The question is important for organizations to consider as they move increasingly to online information for the public. The article describes problems that language-impaired users may face in getting information from the Internet. They focus on language disorders resulting from Alzheimer's disease and brain damage (e.g., a stroke). The article highlights the mental and language skills that are involved in using the Internet 相似文献
14.
《Digital Communications & Networks》2022,8(6):868-876
At present, 5G network technology is being applied to various social network modes, and it can provide technical and traffic support for social networks. Potential friendship discovery technology in 5G-enabled social networks is beneficial for users to make potential friends and expand their range of activities and social hierarchy, which is highly sought after in today's social networks and has great economic and application value. However, the sparsity of the dominant user association dataset in 5G-enabled social networks and the limitations of traditional collaborative filtering algorithms are two major challenges for the friend recommendation problem. Therefore, in order to overcome these problems regarding previous models, we propose a Hybrid Ensemble Multiple Collaborative Filtering Model (HEMCF) for discovering potential buddy relationships. The HEMCF model draws on a special autoencoder method that can effectively exploit the association matrix between friends and additional information to extract a hidden representation of users containing global structural information. Then, it uses the random walk-based graph embedding algorithm DeepWalk to extract another hidden representation of users in the buddy network containing local structural information. Finally, in the output module, the HEMCF model stacks and multiplies the two types of hidden representations of users to ensure that the information mentioned above is concentrated in the final output to generate the final prediction value. The magnitude of the prediction value represents the probability of the users being friends, with larger values representing a high probability of the two users being friends, and vice versa. Experimental results show that the proposed method boosts the accuracy of the relationship prediction over baselines on 3 real-world public datasets dramatically. 相似文献
15.
Web service discovery was the key problem in the study of the Web application.Multiple user community relations between each other on the result of Web service discovery was studied,the user community relations was decomposed into predilection relation,clustering and trust relation,while describing the methods of formalization.Based on the formalized community relations a differential service discovery strategy was put forward,a Web service discovery system framework was built,which can retrieve or recommend candidate Web services step by step to produce the final result according to the “from strong to weak” principle in terms of the correlation between Web service and the user interest background.The experimental results show that proposed method can effectively improve the efficiency and precision of Web service discovery. 相似文献
16.
In this paper we present an aspect-oriented approach to define the navigational concern of Web business processes. Web business processes introduce important modifications into the navigational space and structure of e-commerce applications. From an aspect-oriented perspective, Web business processes are crosscutting concerns, whose elements are scattered and tangled with the rest of navigational elements of e-commerce systems. Web business process modeling lacks modularity, decreasing the system comprehensibility and adaptability. In this paper, we present an aspect-oriented approach to model reusable Web business process. This paper also states Web engineering approaches must be aware of the concepts and techniques introduced by the Aspect Oriented community. 相似文献
17.
基于页面请求的DDOS攻击比传统的海量报文攻击更具杀伤性且检测困难。传统的检测方法包括:特征检测,利用隐马尔科夫模型分析Web页面访问行为。相比明文解析与数学估计,提出了分析用户浏览行为相似性界定HTTP-GetFlood攻击,将用户浏览页面切换与浏览时间的比值表示用户浏览行为,不同用户利用这个表达式计算的浏览行为值很难相等。DDOS攻击爆发时,肉机行为高度一致,导致浏览行为值重复率偏高,当高于一定阀值时判断遭受DDOS攻击。 相似文献
18.
Wireless Personal Communications - Web usage mining is a knowledge discovery technique where a data analyst can discover useful information from the web users’ data. Web contains billions of... 相似文献
19.
随着Internet上Web服务的快速增长,准确、高效地发现Web服务已经是Web服务技术中的难点和关键问题。文章提出基于用户兴趣的Web服务发现方法。首先介绍了对用户兴趣进行挖掘和建模的方法,然后论述了Web服务描述文档和用户兴趣特征之间的相关性分析过程,为得出满足用户兴趣的Web服务发现结果提供了一条可行的路径,提高了Web服务发现的准确度。 相似文献