ASPASIA Knowledge Systems - Knowledge Management / Knowledge Sharing

Knowledge Management & Sharing / Information Retrieval / Data Modelling / Data Bases / Decision Support / Multi-Agent Systems / Recommender Systems / Object Filtering / Clustering / Classification

"System Framework for Generating Open Development Platforms for Web-Service Enabled Applications Using Semantic Web Technologies, Distributed Decision Support Units and Multi-Agent-Systems – INFRAWEBS II"

H. Joachim Nern, G. Agre, T. Atanasova, J. Saarela

WSEAS Transactions on INFORMATION SCIENCE & APPLICATIONS, Issue 1, Vol 1, July 2004, ISSN:1790-0832

keywords: Semantic Web, Web-Services, Ontologies, Distributed Decision Making, Development platforms, Multi-Agent Systems

Abstract The paper introduces a system framework, which enables software and Service providers to generate and establish open and extensible development platforms for Web Service enabled applications. Within the system design several non-functional aspects are addressed and covered by applying a set of powerful (AI-) methodologies: interoperability and adaptability are achieved by exchanging semantically enriched Services among platform partners within P2P environments. The machine processability of ontology based structures in conjunction with "closed loop" approaches in the Service operation, maintenance and feedback cycle (QoS, full life cycle) ensures implicitly a high degree of stable adaptability and scalability, as well as the establishment of extensible platform structures (self-organizing structures). Service composability is guaranteed by applying open standards. To achieve a customizable modular development environment “experience” feedback patterns in distributed rule and case bases (CBR) are archived, evaluated and utilized within distributed decision making processes.

"Distributed Decision Support Embedded in Semantic Web Service Development and Maintenance Environments for Intelligent e-Business Applications"

"INFRAWEBS - Intelligent Framework for Networked Businesses and Governments Using Semantic Web Services and Multi-Agent Systems"

NERN H.-J., ATANASOVA T., AGRE G., SAARELA J.
7th WSEAS International Conference on COMPUTERS, WSEAS, Corfu, Greece, 2003

Key-Words: Semantic Web, Web Services, Ontologies, Semantic Based Collaboration, Semantic Web Unit, Case-Based Reasoning, Knowledge Brokering, Intelligent Knowledge Management, Multi-Agent System

Abstract: In this paper the conceptualization and design of an ICT framework is proposed and discussed - an intelligent framework for supporting organizational networking and the effective sharing of resources. The challenges for this are to:
(i) provide a comfortable and "easy to use" but effective knowledge brokering environment,
(ii) used within a decision supported flexible and highly dynamic collaboration platform, and
(iii) supplemented with add-ons to fulfil security and privacy requirements.
These objectives are achieved by developing software components consisting of:
(i) ontology based and decision supported knowledge management structures applicable to different entities, and
(ii) an interoperability environment realized as a dynamic and reconfigurable platform using semantic web services and multi agent systems,
(iii) supplemented with semantic web service and agent related privacy and security modules.
The framework mirrors the modern and future oriented way of ubiquitous computing, where computers will be relegated to the background of our lives and everybody may be at the same time client, broker and, provider of services.

"New Approach for Online Case-Based Reasoning WEB-Services in Knowledge Management Applications"

Nern H.-J., Pautzke F., Atanasova T.
Euromedia 2003, Plymouth, United Kingdom, April 14-16, 2003

Abstract: Due to the fact that nowadays new knowledge creation and generation is rapidly accelerating, the development of efficient tools and systems for knowledge management and reasoning is essential. A main application field in the wide area of knowledge management is designated to decision support systems - systems which provide comprehensive support in gathering, organisation, refinement and distribution of knowledge.

The initial point for presenting this paper are the results of an European project the authors have been involved in [1]. Consequently and on the basis of this former RTD actions [2, 3, 4] the authors present in this paper an enhancement of a specific type of decision support methodology - a new approach in Case Based Reasoning (CBR) for the general application in WEB-Services. The main realisation trait of this new system is the use of a closed loop approach considering and taking into account statistically evaluated user and experience feedback information implicitly. This closed loop structure assures a stable dynamically growing of the internal case bases and optimal adaptation to the user resp. systems needs.

"Multi-Agent Approach for Task Related Decision Supported Information Retrieval"

Abstract: This paper presents an approach for building research models accessing distributed and heterogene-ous knowledge pools using multi-agent information retrieval. Accordingly task-oriented scenarios for intelli-gent information retrieval are discussed. Searching and browsing activities range from a well-defined search for a specific document to a non-specific task to estimate which kind of information is available. An algorithm for information retrieval is proposed that provides decision support to optimise the search results. The system uses metadata information schemes to provide services in refining user queries to focus a search, in automatically routing queries to relevant servers and in clustering related items.

Key-Words: multi-agent system, decision support, knowledge network, metadata description schemes, information retrieval, distributed and heterogeneous knowledge pools

"Realisation of a Task Oriented Recommender System applied to XPS-Bases Research Networking"

Nern H.-J. et al
Euromedia 2001, 18-20^th April 2001, Valencia, Spain Euromedia 2001

Abstract:The realisation of a task oriented Decision Support System (DSS) used within an XPS-based research network is introduced in this paper. The knowledge (information) retrieval process is instantiated and supervised by a recommender system - supporting the user resp. researcher in determining and classifying his information requirements related to his research task. The retrieval procedure is undertaken stepwise in an iterative way, so that the user is conducted in optimising his task description and accordingly optimising his information retrieval.

The recommender system provides a multistage access cycle for converging the searching and retrieval process based on evaluation of feedback information. The system extracts out of the user’s (researcher’s) predefined task characteristically knowledge patterns and accordingly recommends relevant information objects. These objects are automatically rated with respect to their task related relevancy.

KEYWORDS: Intelligent information retrieval, decision support systems, recommender systems, adaptive knowledge management, document clustering, object classification

"Filtering and Classification of Meta-Data-Information Objects for Knowledge Sharing within an AI-Based Research Network"

Nern H.-J. et al
Euromedia 2001, 18-20^th April 2001, Valencia, Spain Euromedia 2001

ABSTRACT: In this paper the object (document) filtering and classification part of knowledge shared research network is described. As knowledge representation for the objects a modified (extended) version of the Dublin Core is used. The filtering and precleaning process results in a struc-tured feature term vector consisting of weighted object terms. The preparation transaction is divided into 3 stages: a rough pre-cleaning, a (word) stem oriented filtering and a normalisation resp. unification of the term vector. The filtering and normalisation is realised as a fuzzy matching procedure. Subsequently the generated feature vector undergoes a classification process which is performed by using statistical similarity approaches for determining relationship functions among the objects themselves and pretended subject tem vectors, to which the objects are allocated.

KEYWORDS: knowledge sharing systems, meta-data-information schemes, object classification, document clustering, information retrieval, content management

Nern H.-J., Grancharova, A. et al
Euromedia 2000 , May 8-10, 2000, Antwerp, Belgium

Abstract: Due to the increasing demand of globally distributed information within complex researching and developing there is a need for intelligent self adapting global research and development networks, especially under GroupWare aspects. It is proposed to design and establish a framework (shell) for Global (international) Knowledge Shared Research Network. It should be noted that this activity is being carried out under the EU-funded RTD project "Knowledge Shared XPS-Based Research Network Using Multi-Agent-Systems (KNIXMAS)". The entire network (Fig.1) is designed as a global network (via Internet) consisting of Local (national) Research and Working Networks LRWN (via Intranet and/or Internet).

This framework consists of co-operating expertsystems (Intelligent Base Systems - IBS) with decision support modules for research management. The communication, especially the knowledge retrieval (exchange) and requests (Intelligent Knowledge Acquisition) between the LRWNs is provided by applying Multi-Agent- Structures. The centre of each LRWN is the special designed co-operating IBS with coordination capability consisting of a Local and a Global Multi-Agent-Environment LMAE/GMAE, an Intelligent Knowledge Broker IKB, Supervisor Expertsystem SXPS, Dynamical Knowledge- and Data Base DKDB, Decision-Support-Unit DSU, Agent-Communication-Coordination-Unit ACCU and an Administrator-Interface ADI.

On the local side the IBS is connected to several local Work-Research-Spaces WRS established by the individual users (researchers). In the LRWNs the knowledge information exchange is either realized as direct Intranet connections and/or especially information retrieval agents with feedback capabilities (CORBA-standards) to fulfil the groupware aspect. The communication models of the different agents are oriented on CORBA-standards - allowing the use of different multi-agent-environments on heterogeneous platforms. The proposed framework-tool (shell) is universally applicable due to its flexible modular structure realized by using well known system standards. Also, the contents of the flexible databases is transferable and interchangeable. Within KNIXMAS-project it is aimed to realize the local research networks and workingspaces as Intranets in 5 different European countries.

"An Intelligent CORBA-Based Software System for Management of Distributed Computer Applications"

Abstract: In this paper, an intelligent CORBA-based software system is developed that allows building and management of distributed software objects. The intelligence of the system consists of automatically analyzing results obtained by running distributed objects that represent executable program codes. The analysis is performed by a decision making program module which considers results produced by the distributed objects as different alternatives to solve the user-defined task and rates them from best to worst by applying a fuzzy approach. The developed system is used to solve an example that is to select the best control alternative for a chemical reactor. The obtained results show that the rating of alternatives depends on the type of optimality criteria chosen to estimate them and the values assigned to the weighting coefficients for these criteria. It is known that the Common Object Request Broker Architecture (CORBA) is developed by the Object Management Group [1] to address the complexity and high cost of creating distributed object applications. CORBA uses an object-oriented approach for generating software components that can be reused and shared between applications and it automates many common network programming tasks. In this paper, an intelligent CORBA-based software system is developed that allows building and management of distributed software objects as well as analyzing results from running objects that represent executable program codes. This system is accomplished by using VisiBroker for C++ [2] that is a complete implementation of the CORBA 2.0 specification.

Keywords: CORBA, Intelligent distributed object systems, Decision making.

Nern H.-J., Atanasova T. et al
Proc. of 4^th World Multiconference on Circuits, Systems, Communications & Computers (CSCC 2000), Athen, Greece, July 10-15, 2000,

"Distributed Heterogeneous Knowledge Data Base for Control System Design: Multiagent Development and Support"

Nern H.-J., Atanasova T. et al
IEEE-ISIC 2000, Int. Symposium on Intelligent Control, Rio, Patras, Greece, 17-19 July, 2000

"Realisation of a Web Based Team-Workflow, Authoring and Publishing Application"

Nern, H.-J.; Stummer, C. et al
Enable 99 Conference - Enabling Network Based Learning, Espoo, Finland, June 2-5 1999

"Realisation of a Workflow and Redaction System Using MIIS and Oracle Database",

Nern, H.-J., Stummer C. et al
Euromedia 99, Conference, SCS Europe, Munic, April 23-26^th, 1999

Nern H.-J.; Dziech A.; et al
Int. Conference on Systems, Signals, Control, Computers SSCC’98, Durban, South Africa, September 22-24, 1998

Nern, H.-J.; Stummer, C.; et al
Proc. of Euromedia 96, London, UK, Dec. 19-21, 1996, pp. 299-302