Towards Ontology Driven Semantic Conflicts Detection in Web services at Message Level

Introduction

Web services have received an extensive attention because their ability to mitigate the integration problem of distributed applications (Williams, 2003). As a consequence, Web services have been considered as very important elements to create distributed systems on the Web (Boissel-Dallier, Lorré, & Benaben, 2009; Kopecky, Vitvar, Bournez, & Farrell, 2007). From Web service design perspective, Web services are designed to be self-contained and self- described (Breslin, O'Sullivan, Passant, & Vasiliu, 2010; Jun, 2003; Wang, Huang, Qu, & Xie, 2004). Self-contained means that Web services are Web enabled application (J. Yang, 2003), and they are having the necessary contents to enable existing systems without writing any line of code (Endrei et al., 2004). Self-described means that the Web service interface is completely and explicitly describes the Web service schema, operations, and the input and output parameters (Rosen, Lublinsky, Smith, & Balcer, 2008).

Even though, one of the major promises of Web service is to provide semantic interoperability between distributed systems, semantic conflicts are still standing on the way of exchanging the messages seamlessly. This is due to data heterogeneity between the output and the input of message sender and message receiver respectively. In fact this heterogeneity occurs at different levels (data or schema level), and appears in different forms (types) (e.g. precision, schema isomorphism, etc.). Thus, the all conflict types at schema and data level should be considered in order to establish semantic interoperability.

From our perspective, in order to provide an effective approach to solve this problem, the approach should consider the following two different points:

• •

  It should be able to accurately detect the conflict type and the reason that causes this particular conflict type.

• •

  It should be able to provide the appropriate and the efficient solution for the identified conflict type.

Regarding to the second points, a noticeable amount of work has been proposed with the aim of solving semantic conflict between heterogeneous Web services. On the other hand, a close look at the literature shows that, semantic conflict detection is still in the early stage, which require further work to be done to reach its maturity.

Nevertheless, the scope of this paper is focusing on semantic conflicts detection only, which is defined in this work as the process of detecting and identifying the type of semantic conflicts and revealing out the related cause for every detected conflict. The main purpose of our approach is to advance the semantic conflicts resolution process; since the exact conflict types will be given to the resolution technique; and thus, only the necessary solution will be applied to solve the given conflict types.

The rest of this paper is organized as follows. The challenges of web services and semantic conflicts detection are presented in the next section. The section after it briefly reviews the related work and illustrates semantic conflicts detection gap. The illustrative example of this work is described in the section following that. The proposed approach is presented in the section after that. The last section concludes this work and provides some future work directions.

Challenges Of Web Services And Semantic Conflicts Detection

In this section we briefly introduce the semantic conflicts challenges in Web service context, and then expose out the challenges of semantic conflicts detection. These challenges are categorized as follows: