欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
如何快速、准确和高效地发现满足用户需求的、veb服务已成为制约服务发展的瓶颈之一。该文针对现有web服务发现机制中存在的效率低下和查准率不高的两个主要问题，提出了一个基于核BatchSOM神经网络聚类优化的语义web服务发现框架。该框架分别在服务表示阶段引入WordNet和隐含语义索引技术对web服务进行语义扩展和概念语义空间降维；在服务分类阶段利用核机学习理论改进一类适用于wleb服务分类的核BatchSOM神经网络算法；在服务匹配阶段提出一种基于核余弦相似性测度的web服务匹配算法。最后，真实、veb服务数据集上的实验结果验证了所提出方法的可行性。
英文摘要：
With the rapid growth and wide application of Web services, the research on how to accurately, efficiently and rapidly find the desired Web services has become a challenging subject. In order to improve the efficiency and precision for Web service discovery, a semantic Web services discovery framework based on Kernel Batch SOM neural network is proposed. Firstly, by introducing the WordNet and Latent Semantic Index （LSI） into the VSM lexical vectors to extend semantics and reduce the dimension, the resulting VSM semantic vectors can well describe Web services＇ true semantic characterization; Secondly, by using the kernel trick to modify Regular Batch SOM＇s weight updating rule, a kernel Batch SOM neural network is proposed to cluster Web services automatically; Thirdly, a kernel Cosine-based similarity matching mechanism is presented to well estimate the similarity of Web services. Finally, the experiments performed on the real-world Web services collection demonstrate the feasibility of the proposed approaches.
同期刊论文项目
同项目期刊论文
期刊信息
位置：