Concept Convergence in Empirical Domains
How to achieve shared meaning is a significant issue when more than one intelligent agent is involved in the same domain. We define the task of concept convergence, by which intelligent agents can achieve a shared, agreed-upon meaning of a concept (restricted to empirical domains). For this purpose we present a framework that, integrating computational argumentation and inductive concept learning, allows a pair of agents to (1) learn a concept in an empirical domain, (2) argue about the concept’s meaning, and (3) reach a shared agreed-upon concept definition. We apply this framework to marine sponges, a biological domain where the actual definitions of concepts such as orders, families and species are currently open to discussion. An experimental evaluation on marine sponges shows that concept convergence is achieved, within a reasonable number of interchanged arguments, and reaching short and accurate definitions (with respect to precision and recall).
Unable to display preview. Download preview PDF.
- 2.Hirsh, H.: Incremental version-space merging: a general framework for concept learning. PhD thesis, Stanford University, Stanford, CA, USA (1989)Google Scholar
- 3.Kalfoglou, Y., Schorlemmer, M.: Ontology mapping: The state of the art. In: Kalfoglou, Y., Schorlemmer, M., Sheth, A., Staab, S., Uschold, M. (eds.) Semantic Interoperability and Integration, Dagstuhl Seminar Proceedings, Dagstuhl, Germany, vol. 04391 (2005)Google Scholar
- 5.Ontañón, S., Dellunde, P., Godo, L., Plaza, E.: Towards a logical model of induction from examples and communication. In: Proceedings of the 13th International Conference of the Catalan Association for Artificial Intelligence. Frontiers in Artificial Intelligence. IOS Press, Amsterdam (in press, 2010)Google Scholar
- 6.Ontañón, S., Plaza, E.: An argumentation-based framework for deliberation in multi-agent systems. In: Rahwan, I., Parsons, S., Reed, C. (eds.) ArgMAS 2007. LNCS (LNAI), vol. 4946, pp. 178–196. Springer, Heidelberg (2008)Google Scholar
- 7.Ontañón, S., Plaza, E.: Multiagent inductive learning: an argumentation-based approach. In: ICML 2010. Omnipress (2010), http://www.icml2010.org/papers/284.pdf
- 8.Provost, F.J., Hennessy, D.: Scaling up: Distributed machine learning with cooperation. In: Proc. 13th AAAI Conference, pp. 74–79. AAAI Press, Menlo Park (1996)Google Scholar