计算机科学与探索 ›› 2015, Vol. 9 ›› Issue (6): 682-691.DOI: 10.3778/j.issn.1673-9418.1410012

• 系统软件与软件工程 • 上一篇    下一篇

交互式Web服务语义增强

苏  鹏1,2,冯志勇1,2,陈世展1,2+   

  1. 1. 天津大学 计算机科学与技术学院,天津 300072
    2. 天津市认知计算与应用重点实验室,天津 300072
  • 出版日期:2015-06-01 发布日期:2015-06-04

Semantic Enhancements of Web Service with Interaction

SU Peng1,2, FENG Zhiyong 1,2, CHEN Shizhan1,2+   

  1. 1. School of Computer Science and Technology, Tianjin University, Tianjin 300072, China
    2. Tianjin Key Laboratory of Cognitive Computing and Application, Tianjin 300072, China
  • Online:2015-06-01 Published:2015-06-04

摘要: Web服务是面向服务体系结构(service-oriented architecture,SOA)的解决方案的关键技术之一。如何提高Web服务标注质量变得越来越重要。传统的标注方式普遍存在标注质量低和需要人工大量干预的缺点,同时由于本体规模的限制,很多Web服务关键字无法找到相应的概念进行标注。基于用户交互的Web服务标注系统包括阈值选择算法、本体扩充算法和交互标注方法。系统根据反馈信息自动修改标注结果,并记录修改结果,对未来出现的相似问题进行自动处理。为最大化反馈信息的影响,使用特定指标选择标注结果,并由用户进行确认。对于当前本体概念不够完善的问题,使用学习本地本体和在线本体查询两种方法来扩充本体。最后,介绍了相关实验,实验结果表明,经过与用户的交互,标注质量有了显著提高。

关键词: Web服务, 语义标注, 本体, 用户交互

Abstract: Web service is one of the key technologies of SOA (service-oriented architecture) solution. How to improve the quality of annotations of Web services is becoming more and more important. Traditional annotation methods commonly have the drawbacks of low annotation quality and lots of manual intervention. For the limitation of ontology size, many keywords cannot be annotated with relevant concepts. The Web service annotation system based on user interaction includes a threshold selection algorithm with user?s feedback, an algorithm to expand the corresponding ontology and an annotation method with user?s interaction. The system automatically modifies relevant annotations with the feedbacks and records the modification results for relevant annotations in the future. In order to maximize the influence of feedbacks, some indicators with which the most informative annotation results will be chosen for confirmation are defined. For the shortage of ontology concepts, two methods for expanding the ontology are proposed, including learning a local ontology and the inquiry of online ontology. At last, this paper introduces the experiments which show that the quality of annotation is significantly improved after the interaction.

Key words: Web service, semantic annotation, ontology, user interaction