Skip to main content
Log in

PADUA: a protocol for argumentation dialogue using association rules

  • Published:
Artificial Intelligence and Law Aims and scope Submit manuscript

Abstract

We describe PADUA, a protocol designed to support two agents debating a classification by offering arguments based on association rules mined from individual datasets. We motivate the style of argumentation supported by PADUA, and describe the protocol. We discuss the strategies and tactics that can be employed by agents participating in a PADUA dialogue. PADUA is applied to a typical problem in the classification of routine claims for a hypothetical welfare benefit. We particularly address the problems that arise from the extensive number of misclassified examples typically found in such domains, where the high error rate is a widely recognised problem. We give examples of the use of PADUA in this domain, and explore in particular the effect of intermediate predicates. We have also done a large scale evaluation designed to test the effectiveness of using PADUA to detect misclassified examples, and to provide a comparison with other classification systems.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. Getting it right: improving decision-making and appeals in social security benefits. Committee of Public Accounts. London: TSO, 2104 (House of Commons papers, session 2003/04; HC406).

  2. Black swans (Cygnus atratus) are native to the Southern hemisphere.

  3. In some cases legal intermediate predicates are used for convenience, but in others there does seem to be no truth functional relationship between the factors considered in applying the concept: the example in Lindahl and Odelstad (2005) is “living together as husband and wife”. The distinction between these different kinds of intermediate concept is also discussed in Chorley and Bench-Capon (2005).

  4. Our use, which is standard English, should not be confused with the very particular meaning given to the term “strategy” in Game Theory.

  5. Since the complete dataset used in Mitchell (1997) is not publicly available, Chorley and Bench-Capon (2005) used subset of this data collected from various published papers.

  6. We have generated 250 cases presenting female candidates whose age is between 60 and 64 years and should classify as entitled to the benefits (the agent using DS1 may misclassify these cases as not entitled). We also have generated 250 cases presenting candidates from the merchant navy or diplomatic services and should therefore classify as not entitled to benefits (the agent using DS2 may misclassify these as entitled).

  7. The difference in the accuracy scored when the proponent is using DS1 or DS2.

References

  • Agrawal R, Imielinski T, Swami AN (1993) Mining association rules between sets of items in large databases. In: Proceedings of international conference on management of data, (ACM SIGMOD 93). Washington, pp 217–216

  • Aleven V (1997) Teaching case based argumentation through an example and models. PhD thesis, University of Pittsburgh, Pittsburgh, PA, USA

  • Amgoud L, Maudet N (2002) Strategical considerations for argumentative agents (preliminary report). In: Proceedings of 9th international workshop on non-monotonic reasoning (NMR’02). Toulouse, France, pp 409–417. Special session on argument, dialogue, decision

  • Amgoud L, Parsons S (2001) Agent dialogues with conflicting preferences. In: Proceedings of 8th international workshop on agent theories, architectures and languages. Seattle, Washington, pp 1–15

  • Amgoud L, Belabbès S, Prade H (2006) A formal general setting for dialogue protocols. In: Proceedings of AIMSA’06 12th international conference on artificial intelligence: methodology, systems, applications. Varna, Bulgaria, pp 14–15

  • Ashley KD (1990) Modeling legal argument. MIT Press, Cambridge

    Google Scholar 

  • Ashley KD, Brüninghaus S (2003a) A predictive role for intermediate legal concepts. In: Proceedings of Jurix 2003. IOS Press, Amsterdam, pp 153–162

  • Ashley KD, Brüninghaus S (2003b) A predictive role for intermediate legal concepts. In: Proceedings of Jurix 2003. IOS Press, Amsterdam, pp 153–162

  • Atkinson K, Bench-Capon TJM (2005) Legal case-based reasoning as practical reasoning. Artif Intell Law 13(1):93–131

    Article  Google Scholar 

  • Bench-Capon TJM (1991) Knowledge based systems applied to law: a framework for discussion. In: Bench-Capon TJM (ed) Knowledge based systems and legal applications. Academic Press, London, pp 329–342

    Google Scholar 

  • Bench-Capon TJM (1993) Neural nets and open texture. In: 4th international conference on AI and law. ACM Press, Amsterdam, pp 292–297

  • Chorley A, Bench-Capon TJM (2005) AGATHA: using heuristic search to automate the construction of case law theories. Artif Intell Law 13(1):9–51

    Article  Google Scholar 

  • Coenen F, Leng P (2005) Obtaining best parameter values for accurate classification. In: Proceedings of ICDM’05. IEEE, pp 597–600

  • Coenen FP, Leng P, Goulbourne G (2004a) Tree structures for mining association rules. J Data Min Knowl Discov 8(1):25–51

    Article  MathSciNet  Google Scholar 

  • Coenen F, Leng P, Ahmed S (2004b) Data structures for association rule mining: T-trees and P-trees. IEEE Trans Data Knowl Eng 16(6):774–778

    Article  Google Scholar 

  • Coenen F, Leng P, Goulbourne G (2004c) Tree structures for mining association rules. J Data Min Knowl Discov l8(1):25–51

    Article  MathSciNet  Google Scholar 

  • Coenen F, Leng P, Zhang L (2005) Threshold tuning for improved classification association rule mining. In: Proceedings of PAKD’05, LNAI3158. Springer, pp 216–225

  • Goulbourne G, Coenen FP, Leng P (1999) Algorithms for computing association rules using a partial- support tree. In: Proceedings of ES99. Springer, London, pp 142–147

  • Groothius M, Svensson J (2000) Expert system support and juridical quality. In: Proceedings of Jurix 2000, 1–10. IOS Press, Amsterda

  • Kakas AC, Maudet N, Moraitis P (2005) Flexible agent dialogue strategies and societal communication protocols. In: 2nd international workshop on argument in multi agent systems (ArgMAS,’04). Springer LNCS 3366

  • Li W, Han J, Pei J (2001) CMAR: accurate and efficient classification based on multiple class-association rules. In: Proceedings of international conference on data mining (ICDM`01). pp 369–376

  • Lindahl L, Odelstad J (2005) Normative positions within an algebraic approach to normative systems. J Appl Log 17(2):63–91

    Google Scholar 

  • Liu B, Hsu W, Ma Y (1998) Integrating classification and association rule mining. In: Proceedings of international conference on knowledge discovery in databases (KDD‘98). New York, AAAI, pp 80–86

  • Mackenzie J (1979) Question-begging in non-cumulative systems. J Philosoph Log 8:127–133

    MathSciNet  Google Scholar 

  • Minsky M, Papert S (1969) Perceptrons: an introduction to computational geometry. MIT Press, Cambridge

    MATH  Google Scholar 

  • Mitchell TM (1997) Machine learning. McGraw-Hill, NY

    MATH  Google Scholar 

  • Moore D (1993) Dialogue game theory for intelligent tutoring systems. PhD thesis, Leeds Metropolitan University

  • Mozina M, Zabkar J, Bench-Capon T, Bratko I (2005) Argument based machine learning applied to law. J Artif Intell 13(1):53–73

    Google Scholar 

  • National Audit Office (2006) International benchmark of fraud and error in social security systems REPORT BY THE COMPTROLLER AND AUDITOR GENERAL|HC 1387 session 2005–2006|20 July 2006

  • Ontañón S, Plaza E (2006) Arguments and counterexamples in case-based joint deliberation. In: ArgMAS. Hakodate, Japan, pp 36–53

  • Oren N, Norman TJ, Preece A (2006) Loose lips sink ships: a heuristic for argumentation. In: Proceedings of 3rd international workshop on argumentation in multi-agent systems(ArgMAS’06). Hakodate, Japan, pp 121–134

  • Prakken H (2000) On dialogue systems with speech acts, arguments, and counterarguments. In: Proceedings of the 7th European workshop on logic for artificial intelligence. Springer lecture notes in AI, vol 1919. Springer, Berlin, pp 224–238

  • Prakken H (2006) Formal systems for persuasion dialogue. In: The knowledge engineering review, vol 21. pp 163–188

  • Quinlan JR (1998) C4.5: programs for machine learning. Morgan Kaufmann Publishers, Los Altos

    Google Scholar 

  • Rissland EL, Skalak DB, Friedman MT (1993) BankXX. A program to generate argument through case-base research. ICAIL, pp 117–124

  • Walton DN, Krabbe ECW (1995) Commitment in dialogue: basic concepts of interpersonal reasoning. SUNY Press, Albany

    Google Scholar 

  • Wardeh M, Bench-Capon TJM, Coenen FP (2007) PADUA protocol: strategies and tactics. In: Proceedings of ECSQARU, 10th European conference on symbolic and quantitative approaches to reasoning with uncertainty, LNAI 4724. pp 465–476

  • Wardeh M, Bench-Capon TJM, Coenen FP (2008) PISA—pooling information from several agents: multiplayer argumentation from experience. In: Proceedings of AI‘2008. Springer, London, pp 133–146

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Maya Wardeh.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Wardeh, M., Bench-Capon, T. & Coenen, F. PADUA: a protocol for argumentation dialogue using association rules. Artif Intell Law 17, 183–215 (2009). https://doi.org/10.1007/s10506-009-9078-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10506-009-9078-8

Keywords

Navigation