Skip to main content

Advertisement

SpringerLink
Log in

An expert-model and machine learning hybrid approach to predicting human-agent negotiation outcomes in varied data

  • Original Paper
  • Published:
Journal on Multimodal User Interfaces Aims and scope Submit manuscript

Abstract

We present the results of a machine-learning approach to the analysis of several human-agent negotiation studies. By combining expert knowledge of negotiating behavior compiled over a series of empirical studies with neural networks, we show that a hybrid approach to parameter selection yields promise for designing more effective and socially intelligent agents. Specifically, we show that a deep feedforward neural network using a theory-driven three-parameter model can be effective in predicting negotiation outcomes. Furthermore, it outperforms other expert-designed models that use more parameters, as well as those using other techniques (such as linear regression models or boosted decision trees). In a follow-up study, we show that the most successful models change as the dataset size increases and the prediction targets change, and show that boosted decision trees may not be suitable for the negotiation domain. We anticipate these results will have impact for those seeking to combine extensive domain knowledge with more automated approaches in human-computer negotiation. Further, we show that this approach can be a stepping stone from purely exploratory research to targeted human-behavioral experimentation. Through our approach, areas of social artificial intelligence that have historically benefited from expert knowledge and traditional AI approaches can be combined with more recent proven-effective machine learning algorithms.

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

Access this article

Log in via an institution

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

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2

Similar content being viewed by others

Notes

  1. “Wide” datasets have numerous measurable variables, while “deep” ones have many entries. Social dataset often have lots of variables to track but relatively few participants.

  2. Since not all experiments contained the same inputs, statistical differences in the retention rate of data after cleaning (289/485 = 60% vs. 654/769 = 85%) are to be expected.

  3. https://scikit-learn.org/.

  4. https://keras.io/.

  5. A sigmoid serves our purposes to force 0–1 range to form a percentage. A different approach, such as softmax, would force a percentage across all outputs so they must sum to 1. We are looking to get a percentage for each output individually, so we chose the sigmoid.

References

  1. Alto KM, McCullough KM, Levant RF (2018) Who is on craigslist? a novel approach to participant recruitment for masculinities scholarship. Psychol Men Masc 19(2):319

    Article  Google Scholar 

  2. Anthony P, Jennings NR (2003) Developing a bidding agent for multiple heterogeneous auctions. ACM Trans Internet Technol (TOIT) 3(3):185–217

    Article  Google Scholar 

  3. Baarslag T, Kaisers M, Gerding E, Jonker CM, Gratch J (2017) When will negotiation agents be able to represent us? The challenges and opportunities for autonomous negotiators. Sierra, Carles (ed.) In Proceedings of the twenty-sixth international joint conference on artificial intelligence. International joint conferences on artificial intelligence. pp. 4684–4690.

  4. Bengio Y (2013) Deep learning of representations: looking forward. In international conference on statistical language and speech processing (pp. 1–37). Springer, Berlin

  5. Bonnefon JF, Shariff A, Rahwan I (2016) The social dilemma of autonomous vehicles. Science 352(6293):1573–1576

    Article  Google Scholar 

  6. Bordone RC (2000) Teaching interpersonal skills for negotiation and for life. Negot J 16(4):377–385

    Article  Google Scholar 

  7. Castelfranchi C, Falcone R, (1998) Principles of trust for MAS: Cognitive anatomy, social importance, and quantification. In Multi Agent Systems, 1998. Proceedings. International conference on (pp. 72–79). IEEE

  8. Caruana R, Lou Y, Gehrke J, Koch P, Sturm M, Elhadad N (2015) Intelligible models for healthcare: predicting pneumonia risk and hospital 30-day readmission. In proceedings of the 21th ACM SIGKDD international conference on knowledge discovery and data mining (pp. 1721–1730). ACM

  9. Chen T, Guestrin C (2016) Xgboost: a scalable tree boosting system. In proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining (pp. 785–794). ACM

  10. Cohen J (2013) Statistical power analysis for the behavioral sciences. Academic press, Cambridge

    Book  Google Scholar 

  11. De Dreu CK, Koole SL, Steinel W (2000) Unfixing the fixed pie: a motivated information-processing approach to integrative negotiation. J Pers Soc Psychol 79(6):975

    Article  Google Scholar 

  12. de Melo CM, Carnevale P, Gratch J, (2011) The effect of expression of anger and happiness in computer agents on negotiations with humans. In the 10th international conference on autonomous agents and multiagent systems-vol. 3 (pp. 937–944). International foundation for autonomous agents and multiagent systems.

  13. Fox J, Ahn SJ, Janssen JH, Yeykelis L, Segovia KY, Bailenson JN (2015) Avatars versus agents: a meta-analysis quantifying the effect of agency on social influence. Human Comput Interact 30(5):401–432

    Article  Google Scholar 

  14. Gerhart B, Rynes S (1991) Determinants and consequences of salary negotiations by male and female MBA graduates. J Appl Psychol 76(2):256

    Article  Google Scholar 

  15. Goodman B, Flaxman S (2017) European Union regulations on algorithmic decision-making and a “right to explanation.” AI Magazine 38(3):50–57

    Article  Google Scholar 

  16. Gupta T (2018) How i got in the top 1% on Kaggle. towards data science, https://towardsdatascience.com/how-i-got-in-the-top-1-on-kaggle-79ddd7c07f1c

  17. Hubbard R, Ryan PA (2000) The historical growth of statistical significance testing in psychology—And its future prospects. Educ Psychol Measur 60(5):661–681

    Google Scholar 

  18. Johnson E, Gratch J, DeVault D (2017) Towards an autonomous agent that provides automated feedback on students' negotiation skills. In proceedings of the 16th conference on autonomous agents and multiagent systems (pp. 410–418). International foundation for autonomous agents and multiagent systems

  19. Klambauer G, Unterthiner T, Mayr A, Hochreiter S (2017) Self-normalizing neural networks. In: Proceedings of the 31st international conference on neural information processing systems, pp 972–981

  20. Lapuschkin S, Binder A, Montavon G, Muller KR Samek W (2016) Analyzing classifiers: fisher vectors and deep neural networks. In proceedings of the IEEE conference on computer vision and pattern recognition (pp. 2912–2920)

  21. Lucas G, Stratou G, Lieblich S, Gratch J (2016) Trust me: multimodal signals of trustworthiness. In Proceedings of the 18th ACM international conference on multimodal interaction (pp. 5–12). ACM

  22. Mell J, Gratch J (2016) IAGO: interactive arbitration guide online. In AAMAS (pp. 1510–1512)

  23. Mell J, Gratch J (2017) Grumpy and Pinocchio: answering human-agent negotiation questions through realistic agent design. In Proceedings of the 16th conference on autonomous agents and multiagent systems (pp. 401–409). International foundation for autonomous agents and multiagent systems

  24. Mell J, Lucas GM, Gratch J (2018) Welcome to the real world: how agent strategy increases human willingness to deceive. In Proceedings of the 17th international conference on autonomous agents and multi agent systems (pp. 1250–1257). International foundation for autonomous agents and multi agent systems

  25. Mell J, Lucas G, Mozgai S, Boberg J, Artstein R, Gratch J (2018) Towards a repeated negotiating agent that treats people individually: cooperation, social value orientation, & machiavellianism. In Proceedings of the 18th international conference on intelligent virtual agents (pp. 125–132). ACM

  26. Metz Rachel (2018) Google demos Duplex, its AI that sounds exactly like a weird, nice human. Intelligent machines. Downloaded from https://www.technologyreview.com/s/611539/google-demos-duplex-its-ai-that-sounds-exactly-like-a-very-weird-nice-human/

  27. Pennacchiotti M, Popescu AM (2011) A machine learning approach to twitter user classification. In Fifth international AAAI conference on weblogs and social media

  28. Perlich C, Dalessandro B, Raeder T, Stitelman O, Provost F (2014) Machine learning for targeted display advertising: transfer learning in action. Mach Learn 95(1):103–127

    Article  MathSciNet  Google Scholar 

  29. Samek W, Wiegand T, Müller KR (2017) Explainable artificial intelligence: understanding, visualizing and interpreting deep learning models. arXiv:1708.08296

  30. Thompson LL (1991) Information exchange in negotiation. J Exp Soc Psychol 27(2):161–179

    Article  Google Scholar 

  31. Yu B, Kaufmann S, Diermeier D (2008) Classifying party affiliation from political speech. J Inf Technol Politics 5(1):33–48

    Article  Google Scholar 

Download references

Acknowledgements

The authors want to thank our colleague Gale Lucas for the insights on the analytical approach to these results. This research was sponsored by the Army Research Office and was accomplished under Cooperative Agreement Number W911NF-20-2-0053. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Army Research Office or the U.S. Government. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation herein.

Author information

Authors and Affiliations

  1. University of Central Florida, Orlando, FL, USA

    Johnathan Mell

  2. Microsoft, San Francisco, CA, USA

    Markus Beissinger

  3. USC Institute for Creative Technologies, Los Angeles, CA, USA

    Jonathan Gratch

Authors
  1. Johnathan Mell
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Markus Beissinger
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jonathan Gratch
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Johnathan Mell.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Mell, J., Beissinger, M. & Gratch, J. An expert-model and machine learning hybrid approach to predicting human-agent negotiation outcomes in varied data. J Multimodal User Interfaces 15, 215–227 (2021). https://doi.org/10.1007/s12193-021-00368-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12193-021-00368-w

Keyword

Search

Navigation