Decision Tree Analysis in Game Informatics

Konishi, Masato; Okubo, Seiya; Nishino, Tokuzo; Wakatsuki, Mitsuo

doi:10.1007/978-3-319-64051-8_2

Cited by 3 publications

(5 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There have been many studies involving Daihinmin that require (and result in) the creation of strong programs [3]- [4], [6]- [15]; we used the following programs.…”

Section: Daihinmin Programsmentioning

confidence: 99%

“…Programs that play Daihinmin on a computer include those that use heuristics based on human thinking to determine the next hand [12,13] and those that use machine learning such as the Monte Carlo method [6,11]. Other studies of computer Daihinmin include the analysis of the behavior of programs using decision trees [3,4], the extraction of the characteristics of programs using n-gram and classify them [7,8], the characteristics of Daihinmin from program logs [9], and the investigation of the effect of evaluation values in the Monte Carlo method [10,15]. This study presents several measures of the complexity of the card game Daihinmin for the first time to obtain several of its characteristics.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The Characteristics of the Card Game Daihinmin

Wakatsuki

2021

Information Engineering Express

Self Cite

View full text Add to dashboard Cite

Combinatorial game theory has many ways of measuring game complexity. This study presents several measures of complexity for the card game Daihinmin to obtain some of its characteristics. Through computer experiments using programs playing Daihinmin, we collect data on the average number of possible moves and the average length of game in the case of the standard rules and in that of the rules including some local rules such as 11-Back. Then, we calculate some indicators such as the strategic complexity of the game and investigate the characteristics of these indicators. Furthermore, a questionnaire survey is conducted to find out which move a human will play from the initial hand in the specified situation. From moves made by various computer programs in the same situation, we show the difference in playing styles between humans and computers.

show abstract

“…There have been many studies involving Daihinmin that require (and result in) the creation of strong programs [3]- [4], [6]- [15]; we used the following programs.…”

Section: Daihinmin Programsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

The Characteristics of the Card Game Daihinmin

Wakatsuki

2021

Information Engineering Express

Self Cite

View full text Add to dashboard Cite

show abstract

“…The UECda framework is used in many studies on Daihinmin, including those that use analytical methods to determine the characteristics of the computer game [5], propose rating algorithms for imperfect information games [3], and study society of mind theory [6].…”

Section: Lock (Tight Shibari)mentioning

confidence: 99%

“…However, in the Monte Carlo method, the evaluation values assigned to each rank (mibun) are not as important as the ratios of the values between ranks. For example, the evaluation values (5, 4, 3, 2, 1) from a higher rank have the same meaning as (10, 8, 6, 4, 2) and (7,6,5,4,3). That is, the ratio of the values between adjacent ranks is important.…”

Section: Comparison Of Evaluation Valuesmentioning

confidence: 99%