Elsevier

Information Fusion

Volume 21, January 2015, Pages 114-129
Information Fusion

Patterns for context-based knowledge fusion in decision support systems

https://doi.org/10.1016/j.inffus.2013.10.010Get rights and content

Highlights

  • Types and effects of knowledge fusion are distinguished.

  • Knowledge fusion processes taking place in a context aware decision support system are analyzed and investigated.

  • Patterns of context-based knowledge fusion in the context aware decision support system are discovered.

  • Patterns criteria are preservation/changing of structures and autonomies of information/knowledge sources and context.

Abstract

The here presented research focuses on the context-based knowledge fusion patterns. Patterns are discovered based on an analysis and investigation of knowledge fusion processes in a context aware decision support system at the operational stage of the system functioning. At this stage the context-based knowledge fusion processes are manifested around the context. The patterns are generalized in regard to the following three aspects: (1) the effects that the knowledge fusion processes produce in the system; (2) the preservation of internal structures for the context and multiple sources the information/knowledge is fused from; and (3) the preservation of multiple sources and the context autonomies. At that, seven knowledge fusion patterns have been discovered: simple fusion, extension, instantiated fusion, configured fusion, adaptation, flat fusion, and historical fusion.

Introduction

The present research continues the research on knowledge logistics [1]. Several generic knowledge fusion patterns were discovered within the knowledge logistics approach. These patterns generalize knowledge fusion processes occurring at different stages of building and application of a decision support system (DSS). The main objective of the present research is to identify context-based knowledge fusion patterns to generalize knowledge fusion processes occurring in the used DSS at the operational stage of its functioning, i.e., the stage where context aware functions of the DSS come into operation. At this stage the automatic knowledge fusion is supported. Such patterns are supposed to give an insight into the context-based knowledge fusion processes by virtue of the fusion schemes’ typification.

The decision support systems heavily rely upon large volumes of data, information, and knowledge arriving from different sources. Whereas several years ago data fusion used to be the main technology integrating data and information from multiple sources within any DSS, today the focus of data fusion has changed to knowledge fusion. The objective of knowledge fusion is to integrate information and knowledge from multiple sources into some common knowledge that may be used for decision making and problem solving or may provide a better insight and understanding of the situation under consideration [2], [3], [4], [5].

Semantics is the basis to ensure that several information and knowledge sources arrive at the same meaning regarding the situation and information/knowledge being communicated. This explains the fact that ontologies support most efforts in knowledge fusion (e.g., [6], [7], [8], [9], [10]). They provide for a shared and common understanding of some domain that can be communicated across the multiple information and knowledge sources as well as across the sources and DSS; facilitate knowledge sharing and reuse in open and dynamic distributed DSSs; provide means to come to certain conclusions about the contextual data and information; allow entities not designed to work together to interoperate [11].

The present research considers knowledge fusion within a framework of context aware decision support [12]. A central component of the framework is an application ontology. This ontology represents domain and problem solving knowledge fused from different knowledge sources. Thus, the application ontology turns into a knowledge source representing two different types of fused knowledge like domain knowledge and problem solving knowledge.

The context aware DSS built in accordance with the framework is intended to support decisions on involvement of autonomous entities in common activities and scheduling these activities. This DSS was tested by supporting decisions on configuration and planning tasks such as supply chain configuration [13], mobile hospital configuration [14], and planning emergency response actions [15]. In this work the applicability of the research is demonstrated by examples from the emergency management domain. A fire situation is considered as the situation where decisions about fire response plans are made. At that, the DSS solves the problem of organizing an emergency responders’ community and scheduling the responders’ activities.

The main contributions of the present research are the typification of knowledge fusion processes occurring in the context-aware DSSs at the operational stage of their functioning and the generalization of these processes in the form of patterns. The patterns represent the knowledge fusion processes in terms of the effects these processes produce in the context aware DSS and map these effects into the ontology paradigm. Besides, the patterns specify how knowledge fusion impacts on the autonomy and structure of the sources the information/knowledge is fused from.

Discovery of the context-based knowledge fusion patterns starts with the determination where knowledge fusion processes occur in the context aware DSSs. Then processes of knowledge fusion and their effects are investigated. Search for such effects in the DSS that supports the present research is the aim of the following investigation. So, the conceptual framework underlying this DSS is introduced. The context-based knowledge fusion processes are manifested around the context. These processes are revealed and described. Some examples from a fire response scenario accompany these descriptions. At the end of each description an appropriate statement is formulated. The statements create awareness of the aspects to be generalized by the patterns and presents the knowledge fusion result as it appears in the DSS. Finally, the knowledge fusion patterns generalizing each of the statements are introduced for all the revealed knowledge fusion processes.

Section snippets

Knowledge fusion and context aware decision support

This research considers the context awareness as “up-to-date knowledge and run-time understanding of the surrounding environment” [16] [p. 163] (the current situation) by the DSS and the environmental sources. Context aware decision support aims at finding and putting together certain pertinent information allowing for making the informed decisions. The information continuously arrives from multiple diverse heterogeneous sources of data, information, and knowledge located in the environment.

Knowledge fusion processes

The knowledge fusion problem refers to integration of information/knowledge from different sources to obtain new knowledge. The main feature of the knowledge fusion lies in creation of synergetic effect from the integration of information/knowledge. Basically, such effect can be achieved through integration of both tacit and explicit knowledge as well as through their combination. Tacit knowledge integration is embedded in societal activities and interactions. This kind of integration lies

Context aware decision support system

Before the examination of the context-aware DSS for knowledge fusion processes, the conceptual framework underlying this DSS is introduced [12]. The DSS is intended to function in open dynamic environments consisting of a large number of information/knowledge sources. In the framework the context is used to represent the knowledge about the decision situation. The framework relies upon a currently accepted definition of the context: “Context is any information that can be used to characterize

Context-based knowledge fusion in DSS

The processes of knowledge fusion are considered with references to the abstract and operational contexts. These processes are illustrated by examples from the decision support scenario in a fire situation. The knowledge fusion processes identified by the DSS are associated with the numbers assigned to the knowledge fusion processes in Table 1. The results of knowledge fusion are measured in terms of preservation of the internal structures and autonomies of the initial and target knowledge

Results and discussion

In the considered research the knowledge fusion processes of several types were singled out and their possible results were found out. Context-based knowledge fusion processes in the earlier proposed context-aware decision support system for the emergency response application domain were investigated. The aim of the current investigation was to reveal what processes in the system govern the identified results. Based on the statements formulated for the revealed processes the context-based

Acknowledgements

The present research was supported partly by projects funded by Grants 12-07-00298, 13-07-12095, 13-07-13159, 14-07-00345, 14-07-00427 of the Russian Foundation for Basic Research, the project 213 of the research program “Information, control, and intelligent technologies & systems” of the Russian Academy of Sciences (RAS), and the project 2.2 of the Nano- & Information Technologies Branch of RAS. The authors would also like to thank anonymous referees and guest editors of this volume whose

References (52)

  • C. Jonquet et al.

    NCBO resource index: ontology-based search and mining of biomedical resources

    J. Web Semant.

    (2011)
  • A. Smirnov et al.

    Knowledge source network configuration approach to knowledge logistics

    Int. J. Gen. Syst.

    (2003)
  • R. Scherl, D.L. Ulery, Technologies for Army Knowledge Fusion, Final Report, Monmouth University, Computer Science...
  • P. Alun et al.

    Kraft: an agent architecture for knowledge fusion

    Int. J. Coop. Inf. Syst.

    (2001)
  • C.W. Holsapple et al.

    Building blocks for decision support systems

  • A. Smirnov et al.

    Constraint-driven methodology for context-based decision support

    J. Decis. Syst.

    (2005)
  • A. Smirnov et al.

    Ontology engineering for knowledge sharing in supply chains

  • A. Smirnov et al.

    Disaster response based on production network management tasks

    Manage. Res. News

    (2007)
  • A. Smirnov et al.

    Self-organisation of smart environment resources for disaster management and relief

  • F. Burstein et al.

    Context-aware mobile medical emergency management decision support system for safe transportation

  • T. Strang et al.

    A context modeling survey

  • G. Cagalaban et al.

    Context-aware service framework for decision-support applications using ontology-based modeling

  • G. Jakobson, J. Buford, L. Lewis, Towards an architecture for reasoning about complex event-based dynamic situations,...
  • H. Chen, F. Perich, T. Finin, A. Joshi, SOUPA: standard ontology for ubiquitous and pervasive applications, in: Proc....
  • X.H. Wang, T. Gu, D.Q. Zhang, H.K. Pung, Ontology based context modeling and reasoning using OWL, in: Proc. 2nd IEEE...
  • D. Heckman, T. Schwartz, B. Brandherm, M. Von Wilamowitz-Moellendorff, GUMO, the general user model ontology, in: Proc....
  • Cited by (53)

    • On a multisensor knowledge fusion heuristic for the Internet of Things

      2021, Computer Communications
      Citation Excerpt :

      Among the most relevant important applications for knowledge fusion are: According to Smirnov [10] Knowledge fusion is widely used to create and refine knowledge repositories. knowledge fusion systems, usually, compute the correctness of a relation among entities based on the agreement between different extractors and priors [7,8].

    • Knowledge fusion patterns: A survey

      2019, Information Fusion
      Citation Excerpt :

      Intelligent fusion assumes taking into account the semantic content of the sources being fused. The definition above underlies the research on context-based knowledge fusion in decision support systems [61,62]. In this research, multiple knowledge are integrated to produce an ontology-based context model that various knowledge sources use as the basis for problem solving and decision making.

    • Knowledge-based multimodal information fusion for role recognition and situation assessment by using mobile robot

      2019, Information Fusion
      Citation Excerpt :

      It provides a universal understanding of the domains that can be communicated between systems. Various works have been done by high-level knowledge fusion and context awareness [28–31]. By fusing multimodal information, robots can expand their understanding to a decision level.

    View all citing articles on Scopus
    View full text