Knowledge sharing for continuous business engineering based on web intelligence

Modern business trends require from companies continuous business engineering to guarantee a continuous business improvements. Since the Internet has become an easy accessible and popular place business applications the problem of knowledge sharing based on using Web tools and dealing with knowledge representation and processing becomes actual. Existing tools are usually oriented to good structured knowledge sources processing and do not take into account features of model of users’ interests. The paper presents a multi-agent architecture of the systems addressing knowledge logistics as a tool providing for a web-based intelligent service of continuous business engineering. agement (KM) called knowledge logistics (KL) (Smirnov et al. 2002a). KL is based on individual user requirements, available knowledge sources, and current situation analysis. This technology enables EAI in the following aspects: business process integration, application integration, data integration. Proposed in the paper KSNet-approach to knowledge logistics is oriented to “Just-Before-Time” service for intelligent support of CBE. Advanced technologies of intelligent agents, ontology management, constraint satisfaction, profiling, and knowledge fusion underlie the proposed approach. Today, Web services are believed to be the crucial technology for business. Web service can be seen as high-level interfaces through which partners can conduct business operations. The priority modern industrial projects involve Web applications. This is not surprising because business is becoming more Internet-dependent. The integration of Web applications with existing systems is a key driver of EIA (Brown, 2003). Wide spread of modern information technologies, such as World Wide Web and intelligence agents (Huhns and Stephens, 2000), has led to an appearance of a new direction for scientific research and development called "Web intelligence". Web intelligence explores fundamental and practical impacts of AI and advanced Information Technologies on the next generation of Web-empowered systems, services, and environments. The basis for the ongoing Web intelligence research agenda is made up the issues: (i) Web mining that applies data mining techniques to large Web data repositories; (ii) Web-based knowledge processing and management that focuses on developing the semantic Web, the base of this research is ontological knowledge representation; (iii) distributed inference engines that perform automatic reasoning on the Web; (iv) information exchange and knowledge sharing coupled with human-crafted resources that support sustainable knowledge creation (Zhong et al. 2002). The paper is organized as follows. Section 2 elucidates knowledge logistics concerned with ontology approach. Section 3 presents main components of the system “KSNet”. Section 4 describes chosen knowledge sharing model of the system “KSNet”, architecture of the developed research prototype and web agent architecture. Section 5 presents Knowledge Fusion agent features as the most important problem-oriented web agent of the system. Main features of the system that correspond to fundamental capabilities of the intelligent Web’s design and development are presented in conclusion. 2 ONTOLOGY-DRIVEN KNOWLEDGE LOGISTICS Knowledge logistics addresses the problem of acquisition of the right knowledge from distributed sources, its integration and transfer to the right person within the right context, at the right time, for the right purpose. This problem in the approach is considered as a network configuration that includes endusers, loosely coupled knowledge sources, and a set of tools and methods for knowledge processing located in an e-business environment. Such network of loosely coupled sources was referred to as the knowledge source network or “KSNet”. The main principles considered during the development of the proposed approach and a KL system based on it originate from the characteristics of modern “e”-applications. These applications widely use ontologies as a common language for business process / enterprise modelling (Goossenaerts & Pelletier 2001, O’Leary 2000, OILEd 2002, Protégé 2003, Semantic Web 2003). Thus, the approach focuses on utilizing reusable knowledge through shared ontological representations. The application of intelligent agents representing their knowledge via ontologies (Weiss, 2000) was motivated by the need of knowledge logistics systems for flexibility, scalability, and customizability. The multiagent system architecture based on the FIPA Reference Model (FIPA 2002) was chosen as a technological basis for the definition of agents’ properties and functions since it provides standards for heterogeneous interacting agents and agent-based systems, and specifies ontologies and negotiation protocols. As a formal model for knowledge integration the ontology model with the knowledge representation formalism of object-oriented constraint networks was chosen. This allows simplifying the formulation and interpretation of real-world problems which in the areas of engineering, manufacturing, management, etc. are usually presented as constraint satisfaction problems (Smirnov et al. 2002a). The object-oriented constraint networks formalism (Smirnov, 2001) was chosen as the abstract model for ontology representation (Fig. 1). The abstract model based on this notation unifies main concepts of languages, such as standard objectoriented languages with classes, and constraint programming languages. It supports the declarative representation, efficiency of dynamic constraint solving, and problem modelling capability, maintainability, reusability, and extensibility of the object-oriented technology. According to the paradigm the knowledge can be described by classes, attributes, domains, constraints, and methods. This perspective of knowledge representation correlates well with the semantic metadata representation concept being developed under the Semantic Web project (Semantic Web, 2003). Class A Class B Superclasses