search for


Betting strategies for quiz game in question-and-answer system
Journal of the Korean Data & Information Science Society 2018;29:415-24
Published online March 31, 2018
© 2018 Korean Data and Information Science Society.

Jonghwa Na1 · Eunji Hwang2

12Department of Information and Statistics, Chungbuk National University
Correspondence to: Professor, Department of Information and Statistics, Chungbuk National University, Chungbuk 28644, Korea. Email:
Received October 30, 2017; Revised February 26, 2018; Accepted March 2, 2018.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Confrontation between artificial intelligence computer and human is one of great concern in many field. In 2011, IBM’s Watson computer won the quiz battle against human quiz champions. Korea electronics and telecommunications research institute (ETRI) has also developed WiseQA, a question and answering system for confrontation with humans. WiseQA has been trained in preparation for the EBS scholarship quiz. This requires the development of algorithms for rational decision making. In this study, a model was constructed to make a decision based on statistical model for the problem that WiseQA should judge during the quiz showdown (judgment of buzz-in and determination of betting score). We considered three statistical models for the buzz-in strategy: logistic regression, decision tree, and neural network model. Also, the multi-category logit model and proportional odds logit model was constructed as a strategy for determining the betting score. As a result, there was no significant difference in the performance between the three models for the buzz-in strategy. In the betting strategy, it was confirmed that the recommended betting score based on the statistical model was better than the one determined by the human participants.
Keywords : Artificial intelligence, decision tree, logistic regression, multi-category logit model, neural network, proportional odds logit model.