An argumentation framework for merging conflicting knowledge bases

• Authors: Leila Amgoud;Souhila Kaci
• Venue: International Journal of Approximate Reasoning
• Year: 2007

The problem of merging multiple sources of information is central in many information processing areas such as databases integrating problems, multiple criteria decision making, etc. To solve this problem, two kinds of approaches have been proposed. The first category of approaches merges the different bases into a unique consistent base, and the second category, such as argumentation, accepts inconsistency and copes with it. It is well known that priorities are crucial to solve conflicts. Recently, powerful approaches have been proposed to merge multiple sources information where priorities are either explicitly or implicitly associated to information [L. Cholvy, Reasoning about merging information, Handbook of Defeasible Reasoning and Uncertainty Management Systems, vol. 3, 1998, pp. 233-263; S. Konieczny, R. Pino Perez, On the logic of merging, in: Proceedings of the 6th International Conference on Principles of Knowledge Representation and Reasoning (KR'98), Trento, 1998, pp. 488-498; J. Lin, Integration of weighted knowledge bases, Artificial Intelligence 83 (1996) 363-378; J. Lin, A. Mendelzon, Merging databases under constraints, International Journal of Cooperative Information Systems 7(1) (1998) 55-76; N. Rescher, R. Manor, On inference from inconsistent premises, Theory and Decision 1 (1970) 179-219; P.Z. Revesz, On the semantics of theory change: arbitration between old and new information, in: 12th ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Databases, 1993, pp. 71-92; S. Benferhat, D. Dubois, S. Kaci, H. Prade, Possibilistic merging and distance-based fusion of propositional information, Annals of Mathematics and Artificial Intelligence, 34(1-3) (2002) 217-252; S. Benferhat, D. Dubois, H. Prade, M. Williams, A practical approach to fusing and revising prioritized belief bases, in: Proceedings of the 9th Portuguese Conference on Artificial Intelligence (EPIA'99), 1999, pp. 222-236; S. Kaci, Connaissances et Preferences: Representation et fusion en logique possibiliste, These de doctorat, Universite Paul Sabatier, Toulouse, 2002]. In this paper, we present an argumentation framework for solving conflicts which could be applied to conflicts arising between agents in a multi-agent system. We suppose that each agent is represented by a knowledge base and that the different agents are conflicting. We show that the argumentation framework retrieves the results of the merging approaches. Moreover, an argumentation-based approach palliates the limits, due to the drowning problem, of the merging operator when information is pervaded with explicit priorities.