Sharing of Semantically Enhanced Information for the Adaptive Execution of Business Processes

TL;DR

This paper introduces the Semantic Context Space (SCS) Engine, a system designed to facilitate adaptable business processes by storing and exchanging semantically annotated data using a space-based model inspired by Semantic TupleSpace and JavaSpace.

Contribution

The paper presents the design and implementation of the SCS Engine, enabling semantic data storage and retrieval for process adaptation, based on ontologies and space-based operations.

Findings

The SCS Engine supports write, read, and take operations with semantic constraints.

Experimental evaluation confirms the applicability of the SCS Engine for data-driven process adaptation.

The system integrates semantic annotations with space-based data exchange mechanisms.

Abstract

Motivated from the Context Aware Computing, and more particularly from the Data-Driven Process Adaptation approach, we propose the Semantic Context Space (SCS) Engine which aims to facilitate the provision of adaptable business processes. The SCS Engine provides a space which stores semantically annotated data and it is open to other processes, systems, and external sources for information exchange. The specified implementation is inspired from the Semantic TupleSpace and uses the JavaSpace Service of the Jini Framework (changed to Apache River lately) as an underlying basis. The SCS Engine supplies an interface where a client can execute the following operations: (i) write: which inserts in the space available information along with its respective meta-information, (ii) read: which retrieves from the space information which meets specific meta-information constrains, and (iii) take: which retrieves and simultaneously deletes from the space information which meets specific meta-information constrains. In terms of this thesis the available types of meta-information are based on ontologies described in RDFS or WSML. The applicability of the SCS Engine implementation in the context of data-driven process adaptation has been ensured by an experimental evaluation of the provided operations. Ultimately, we refer to open issues which could be addressed to enrich the proposed Engine with additional features.