Abstract
To date, a variety of automated negotiation agents have been created. While each of these agents has been shown to be effective in negotiating with people in specific environments, they lack natural language processing (NLP) methods required to enable real-world types of interactions. In this paper we study how existing agents must be modified to address this limitation. After performing an extensive study of agents’ negotiation with human subjects, we found that simply modifying existing agents to include an NLP module is insufficient to create these agents. Instead the agents’ strategies must be modified to address offers that do not include values for all the discussed issues (as is the case in menu-based interfaces) and consequently issue-by-issue interactions.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The exact functions (and the complete domain) are available in the GENIUS framework that can be freely downloaded from the Internet.
- 2.
- 3.
The state-of-the-art in NLU for dialog systems is sequence classification [12]. We decided against this option because it requires too much labeling effort: while in multi-label classification you only need to label each sentence, in sequence classification you must label each fraction of a sentence. After deciding to use multilabel classifiers, we checked various approaches to multi-label classification [21] and various kinds of base binary classifiers. We found out that the combination of HOMER with Modified Balanced Winnow, described above, had the best performance in terms of both classification accuracy and run-time.
- 4.
We tried more sophisticated features, such as pairs of non-adjacent words, but this didn’t improve performance.
- 5.
Sentence-level accuracy is the number of sentences whose classification was exactly correct (i.e., the set of dialog acts returned by the MLC is identical to the correct set), divided by the total number of sentences. The 72Â % accuracy was calculated using 5-fold cross-validation on the set of 775 tagged sentences. Sentence-level accuracy is the strictest possible performance measure. In other measures, such as precision, recall or F1, the performance of our NLU was higher
References
Baarslag, T., Fujita, K., Gerding, E.H., Hindriks, K., Ito, T., Jennings, N.R., Jonker, C., Kraus, S., Lin, R., Robu, V., Williams, C.R.: Evaluating practical negotiating agents: results and analysis of the 2011 international competition. Artif. Intell. 198, 73–103 (2013)
Bac, M., Raff, H.: Issue-by-issue negotiations: the role of information and time preference. Games Econ. Behav. 13(1), 125–134 (1996)
Busch, L.-A., Horstmann, I.: A comment on issue-by-issue negotiations. Games Econ. Behav. 19(1), 144–148 (1997)
Byde, A., Yearworth, M., Chen, K.-Y., Bartolini, C.: Aut ONA: a system for automated multiple 1–1 negotiation. In: Proceedings of the 2003 IEEE International Conference on Electronic Commerce (CEC), pp. 59–67 (2003)
Carvalho, V.R., Cohen, W.W.: Single-pass online learning: performance, voting schemes and online feature selection. In: Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD’06, pp. 548–553. ACM, New York (2006)
Chen, MK.: Agendas in multi-issue bargaining: when to sweat the small stuff. Technical report, Harvard Department of Economics, Cambridge, November 2002
Coen, M.H.: Design principles for intelligent environments. In: AAAI/IAAI, pp. 547–554 (1998)
Cohen, P.R.: The role of natural language in a multimodal interface. In: Proceedings of the 5th Annual ACM Symposium on User Interface Software and Technology, UIST’92, pp. 143–149. ACM, New York (1992)
Dagan, I., Roth, D., Sammons, M., Zanzotto, F.M.: Recognizing textual entailment: models and applications. Synth. Lect. Hum. Lang. Technol. 6(4), 1–220 (2013)
Dahlbäck, N., Jönsson, A., Ahrenberg, L.: Wizard of Oz studies: why and how. In: Proceedings of the 1st International Conference on Intelligent User Interfaces, IUI’93, pp. 193–200. ACM, New York (1993)
Gal, Y., Kraus, S., Gelfand, M., Khashan, H., Salmon, E.: An adaptive agent for negotiating with people in different cultures. ACM TIST 3(1), 8 (2011)
Hahn, S., Dinarelli, M., Raymond, C., Lefevre, F., Lehnen, P., de Mori, Renato, Moschitti, A., Ney, H., Riccardi, G.: Comparing Stochastic Approaches to Spoken Language Understanding in Multiple Languages. IEEE Transactions on Audio, Speech, and Language Processing, vol.19, no.6, pp.1569–1583 (2011). http://dx.doi.org/10.1109/tasl.2010.2093520
Jonker, C.M., Robu, V., Treur, J.: An agent architecture for multi-attribute negotiation using incomplete preference information. Auton. Agents Multi-Agent Syst. 15(2), 221–252 (2007)
Jurafsky, D., Martin, James H.: Speech and Language Processing, 2nd edn, Prentice Hall (2008). ISBN:0131873210
Katz, R., Kraus, S.: Efficient agents for cliff edge environments with a large set of decision options. In: Proceedings of the 5th International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), pp. 697–704 (2006)
Kelley, J.F.: An empirical methodology for writing user-friendly natural language computer applications. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, CHI’83, pp. 193–196. ACM, New York (1983)
Kenny, P., Hartholt, A., Gratch, J., Swartout, W., Traum, D., Marsella, S., Piepol, D.: Building interactive virtual humans for training environments. In: Proceedings of Interservice/Industry Training, Simulation and Education Conference (I/ITSEC) (2007)
Kraus, S., Lehmann, D.: Designing and building a negotiating automated agent. Comput. Intell. 11(1), 132–171 (1995)
Lin, R., Kraus, S.: Can automated agents proficiently negotiate with humans? CACM 53(1), 78–88 (2010)
Lin, R., Kraus, S., Wilkenfeld, J., Barry, J.: Negotiating with bounded rational agents in environments with incomplete information using an automated agent. Artif. Intell. 172(6–7), 823–851 (2008)
Madjarov, G., Kocev, D., Gjorgjevikj, D., Džeroski, S.: An extensive experimental comparison of methods for multi-label learning. Pattern Recogn. 45(9), 3084–3104 (2012). http://dx.doi.org/10.1016/j.patcog.2012.03.004
Osborne, M.J., Rubinstein, A.: A Course In Game Theory. MIT Press, Cambridge (1994)
Oshrat, Y., Lin, R., Kraus, S.: Facing the challenge of human-agent negotiations via effective general opponent modeling. In: Proceedings of the 8th International Conference on Autonomous Agents and Multiagent Systems (AAMAS) (2009)
Pease, A., Colton, S., Smaill, A., Lee, J.: Semantic negotiation: Modelling ambiguity in dialogue. In: Proceedings of Edilog 2002, the 6th Workshop on the Semantics and Pragmatics of Dialogue, Edinburgh, UK (2002)
Shneiderman, B., Plaisant, C.: Designing the User Interface: Strategies for Effective Human-Computer Interaction, 4th edn. Pearson Addison Wesley, Boston (2004)
Tsoumakas, G., Katakis, I., Vlahavas, I.: Effective and efficient multilabel classification in domains with large number of labels. In: Proceedings ECML/PKDD 2008 Workshop on Mining Multidimensional Data (MMD’08) (2008)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer Japan
About this chapter
Cite this chapter
Zuckerman, I., Segal-Halevi, E., Rosenfeld, A., Kraus, S. (2015). First Steps in Chat-Based Negotiating Agents. In: Fujita, K., Ito, T., Zhang, M., Robu, V. (eds) Next Frontier in Agent-Based Complex Automated Negotiation. Studies in Computational Intelligence, vol 596. Springer, Tokyo. https://doi.org/10.1007/978-4-431-55525-4_6
Download citation
DOI: https://doi.org/10.1007/978-4-431-55525-4_6
Publisher Name: Springer, Tokyo
Print ISBN: 978-4-431-55524-7
Online ISBN: 978-4-431-55525-4
eBook Packages: EngineeringEngineering (R0)