Classification based Parameter Association for Non-redundant Annotation
Xiaocao Hu, Zhiyong Feng, Keman Huang, Shizhan Chen
*
Tianjin Key Laboratory of Cognitive Computing and Application
School of Computer Science and Technology, Tianjin University
Tianjin, China
{huxiaocao, zyfeng, keman.huang, shizhan}@tju.edu.cn
Abstract—Semantic annotation of Web Services can facilitate
the automated service discovery and composition. At present,
however, many solutions suffer from redundant annotations or
imprecise derived annotations. The fundamental task to address
the issue is to find parameters that have same semantics in a
large number of Web Services. This paper proposes a
classification based approach for identifying parameters that are
semantically equivalent, laying the foundation for non-redundant
annotation. Subsequently the paper presents the non-redundant
annotation, in which the parameter space is firstly reduced, and
then semantic annotation is performed on the reduced parameter
space, finally annotation results of the reduced parameter space
are expanded to the original parameter space. To evaluate the
final annotation results, the paper gives a methodology based on
service discovery. The experimental results indicate that the
parameter association approach can achieve outstanding
accuracy in identifying the equivalence between parameters.
Moreover, it is suggested that the non-redundant annotation can
greatly reduce redundant annotations, and can augment the
semantics of Web Services with adequate accuracy, which can
achieve similar performance in service discovery as OWL-S
services do.
Keywords—Web Service; Parameter Association; Classification;
Non-redundant Annotation
I. INTRODUCTION
With the spread of Web Services [1] in various fields, there
is a growing interest in specifying the semantic description of a
service, which facilitates the automated service discovery and
composition. Earlier research of specifying the semantics of
services focuses on manual annotation [2], which poses several
problems. Firstly, the burden of choosing relevant ontologies
lies with the user. Secondly, the potential number of concepts
in a Web Service increases manifold as the service description
grows larger. Thirdly, the ontology used for annotation could
be very large with correspondingly large number of concepts.
As a consequence, it becomes a very tedious task to manually
specify the semantics of services. To resolve the difficulty of
manual annotation, many studies have been carried out on the
automatic semantic annotation of Web Services. A well-
adopted strategy is to associate Web Services, especially inputs
and outputs of the service, with concepts from ontologies.
Most approaches on the automatic semantic annotation of
Web Services can be classified into three main categories,
annotation using schema matching techniques [3, 7-8, 10-11],
annotation using machine learning algorithms [4-5, 9], and
annotation using data-driven workflows [6]. For the first two
categories, the annotation process is performed whenever
parameters of a new Web Service come, even when parameters
of the new service have same semantics with parameters of
services that are already annotated, thus resulting in redundant
annotations, which may be a large number of calculations. For
the third category, annotation information of parameters on one
side of a data link is derived by the other side of the data link,
which is already annotated. In this case, the redundancy is
avoided, however, derived annotations are imprecise and exact
annotations cannot be inferred.
To reduce redundant annotations and guarantee exact
annotations, the main task is to efficiently find parameters that
have same semantics in a large number of Web Services. In
this paper, we propose a classification based approach for
identifying parameters that are semantically equivalent. The
approach lays the foundation for reducing the parameter space
for annotation. Then the semantic annotation is performed on
the reduced parameter space instead of the original parameter
space, thus avoiding redundant annotations. And the annotation
result of a parameter is directly applied to other parameters that
have same semantics, thus guaranteeing exact annotations. In
addition, a methodology is presented to evaluate the annotation
results. The major contributions of our work are summarized as
follows:
x An efficient approach for finding parameters that have
same semantics in different Web Services.
x An overall process of the non-redundant annotation and
a methodology for evaluating the annotation results.
x A detailed experimental evaluation for the parameter
association approach and the non-redundant annotation.
The rest of this paper is organized as follows. Section II
reviews the related work. Section III describes the details of
parameter association approach. Section IV introduces the
overall process of the non-redundant annotation, and presents a
methodology for the evaluation of annotation results. Section V
conducts experiments while Section VI draws the conclusion.
II. R
ELATED WORK
In the field of semantic annotation, many solutions have
been developed. Patil [3] presents a framework, METEOR-S
Web service Annotation Framework (MWSAF), to semi-
automatically annotate WSDL descriptions of the services with
relevant ontologies. Firstly, WSDL files and ontologies are
represented by the same format through two translators.
Secondly, elements in the WSDL files are mapped to concepts
of an ontology, and the best mapping is selected.
2015 IEEE International Conference on Services Computing
978-1-4673-7281-7/15 $31.00 © 2015 IEEE
DOI 10.1109/SCC.2015.98
688
2015 IEEE International Conference on Services Computing
978-1-4673-7281-7/15 $31.00 © 2015 IEEE
DOI 10.1109/SCC.2015.98
688
2015 IEEE International Conference on Services Computing
978-1-4673-7281-7/15 $31.00 © 2015 IEEE
DOI 10.1109/SCC.2015.98
688
2015 IEEE International Conference on Services Computing
978-1-4673-7281-7/15 $31.00 © 2015 IEEE
DOI 10.1109/SCC.2015.98
688
剩余7页未读，继续阅读
资源评论
weixin_38641111
- 粉丝: 1
- 资源: 931
最新资源
- mamba 2 windows （cuda12.4）安装包，包括causal-conv1d，mamba-ssm
- simulink风储调频，风机混合储能电池超级电容储能联合一次调频，系统频率特性如下 电池储能采用双闭环pwm设计，频率控制环节为下垂控制 超级电容采用恒流充放电，降低电压释放存储在电容器中的能量
- OPC用户使用操作使用说明
- 0java实验报告.7z
- 两层液压梯程序，使用三菱FX1N系列PLC，已在设备上正常使用，非常稳定可靠，包含PLC程序
- zb25电气调车参考文档
- C++、基于MFC图像处理系统-2025
- JAVA基础编程练习：涵盖条件语句、循环结构、类和对象、继承多态及异常处理的应用案例集合
- zb45电气调车参考文档
- C++期末大作业-MFC 开发的Windows标准计算器和程序员计算器-2025
- Java编程实践：多位数字解析、文本检索、格式校验及其他基础练习
- python 的虚拟环境 原理
- C++-MFC框架为基础，结合opencv来对图像做处理-2025
- LSSVM，SSA-LSSVM，VMD-LSSVM，VMD-SSA-LSSVM四种算法做短期电力负荷预测，做对比 结果分析-lssvm 均方根误差(RMSE)：0.79172 平均绝对误差（MAE）
- JAVA面向对象编程与异常处理核心技术解析及实际应用
- ZB45电气简化图册1
资源上传下载、课程学习等过程中有任何疑问或建议，欢迎提出宝贵意见哦~我们会及时处理！
点击此处反馈
