1 \documentclass[12pt,
a4paper,notitlepage
]{article
}
3 \usepackage[a4paper,vmargin=
{40mm,
40mm
},hmargin=
{30mm,
30mm
}]{geometry
}
5 %% Použité kódování znaků: obvykle latin2, cp1250 nebo utf8:
6 \usepackage[utf8
]{inputenc}
9 \usepackage[titletoc
]{appendix
}
13 \usepackage[multiple
]{footmisc
}
14 %\usepackage[dont-mess-around]{fnpct}
17 \usepackage{threeparttable
}
18 \usepackage{longtable
}
20 %\usepackage{amsfonts}
22 %\usepackage[lined, ruled, boxed, linesnumbered]{algorithm2e}
25 \usepackage[round
]{natbib
} % sazba pouzite literatury
28 %\usepackage{psgo,array}
29 \usepackage{url
} % sazba URL
31 \usepackage[ps2pdf,unicode
]{hyperref
} % Musí být za všemi ostatními balíčky
34 \title{Style Consensus: Style of Professional Players\\
35 Judged by Strong Players
}
37 \author{Josef~Moud
\v{r
}\'
{i
}k,~Petr~Baudi
\v{s
}\\
38 \small Faculty of Math and Physics, Charles University, Prague, CZ
}
44 GoStyle
\citep{GoStyleWeb
} is a project we founded to study
45 possibilities of computer analysis of databases of Go games. As a
46 part of this project, we conducted (partly manual and partly online)
47 a questionnaire, where we ask experts (strong amateur or professional players)
48 to judge styles of several professional Go players. The experts
49 were asked to judge each professional on four scales traditional
50 Go knowledge. In this
report, we publish these results, along with
51 definitions of styles and acknowledgement of the interviewees. The
52 purpose is to make the data available to the general public.
56 \section{Questionnaire Setup
}
58 We chose a small subset of well known players (mainly from the
20th century) and
59 asked some experts (professional and strong amateur players; see acknowledgement)
60 to evaluate these players using a questionnaire. Initially (first three experts from the
61 acknowledgement) this was done by hand using an e-mail based questionnaire, the rest
62 of the experts were asked using our online questionnaire
\footnote{
63 \url{http://gostyle.j2m.cz/questionare.html
}
66 The experts were asked to value the players on four scales, each ranging from
1 to
10.
71 \begin{tabular
}{|c|c|c|
}
73 \textbf{Style
} &
\textbf{1} &
\textbf{10}\\
\hline
74 Territoriality & Moyo & Territory \\
75 Orthodoxity & Classic & Novel \\
76 Aggressivity& Calm & Fighting \\
77 Thickness & Safe & Shinogi \\
\hline
81 The scales try to reflect
82 some of the traditionally perceived playing styles.
\footnote{
83 Refer to~
\citet{GoGoD:styles
}, or~
\citet{senseis:styles
}.
85 For example, the first scale (
\emph{territoriality
})
86 stresses whether a player prefers safe, yet inherently smaller territory (number
10 on the scale),
87 or roughly sketched large territory (
\emph{moyo
},
1 on the scale), which is however insecure.
88 The meaning of boundary terms in all four scales (maybe except the last one)
89 should be clear to any fairly experienced player.
90 Apart from the table above, the experts had not thus been given any more information regarding the
93 There has been some discussion
\citep[Discussion
]{Moudrik13
}
94 about proper meaning of the last scale and we might want to redefine
95 it in the future. Apart from this, we also plan to
96 rename the Scale of Orthodoxity to Novelty in the further data collection,
97 so that the name of the scale
98 reflects the ``trend'' given by names of the boundaries --- in the questionnaire
99 presented so far, the number
1 was unfortunately
100 assigned to Classic style of game and
10 to Novel style of game.
104 Are given in the Table~
\ref{questionare
} along with standard deviation of individual
105 answers. Mean standard deviation of the answers is
1.302, so we consider the results
108 The following table lists mean value of answers within each scale (to show how
109 are the
1 to
10 scales ``populated'') and pairwise correlations
110 between different styles, using the
111 Pearson's correlation coefficient
\citep{Pearson
}.
115 \begin{tabular
}{|c|c|c|c|c|c|
}
117 \multicolumn{2}{|c|
}{---
} &
118 \multicolumn{4}{|c|
}{\textbf{Pearson's $r$
}} \\
\hline
119 \textbf{Style
} &
\textbf{Mean value
}
120 & Ter.& Orth.& Aggr.& Thick.\\
\hline
122 Territoriality & $
5.762 \pm 2.418$ &
123 1.000 & -
0.574 & -
0.638 &
0.339 \\
124 Orthodoxity & $
5.494 \pm 2.209$ &
125 &
1.000 &
0.730 &
0.105 \\
126 Aggressivity & $
6.679 \pm 2.135$ &
128 Thickness & $
4.954 \pm 1.645$ &
132 \caption[Mean values of styles and their
133 pairwise correlation
]{The mean values of styles (across all the answers) and the pairwise
134 correlation between them.
}
139 % increase table row spacing, adjust to taste
140 \renewcommand{\arraystretch}{1.4}
141 \begin{threeparttable
}
143 \begin{tabular
}{|c|c|c|c|c|c|
}
145 \textbf{\#
} &
\textbf{Player
} &
{\small \textbf{Territoriality
}}\hspace{-
1pt
} &
146 {\small\textbf{Orthodoxity
}}\hspace{-
1pt
} &
147 \hspace{-
1pt
}{\small\textbf{Aggressivity
}}\hspace{-
1pt
} &
148 {\small\textbf{Thickness
}}\hspace{-
1pt
} \\
150 5& Cho Chikun & $
8.4 \pm 1.4$ & $
5.8 \pm 2.5$ & $
7.2 \pm 1.3$ & $
9.2 \pm 0.7$ \\
151 5& Cho U & $
7.2 \pm 1.8$ & $
6.0 \pm 2.0$ & $
6.6 \pm 2.1$ & $
6.4 \pm 1.4$ \\
152 5& Gu Li & $
6.4 \pm 1.2$ & $
7.4 \pm 1.5$ & $
9.0 \pm 0.9$ & $
5.4 \pm 1.9$ \\
153 5& Ishida Yoshio & $
8.6 \pm 1.4$ & $
3.4 \pm 2.2$ & $
3.0 \pm 1.1$ & $
4.2 \pm 1.5$ \\
154 5& Otake Hideo & $
5.4 \pm 1.9$ & $
3.2 \pm 1.6$ & $
3.8 \pm 1.5$ & $
3.4 \pm 1.0$ \\
155 5& Sakata Eio & $
7.6 \pm 1.6$ & $
3.4 \pm 1.6$ & $
8.0 \pm 1.1$ & $
8.2 \pm 1.3$ \\
156 5& Takemiya Masaki & $
1.4 \pm 0.5$ & $
5.0 \pm 2.4$ & $
7.2 \pm 0.7$ & $
2.0 \pm 0.9$ \\
157 5& Yi Ch'ang-ho & $
7.8 \pm 1.7$ & $
5.6 \pm 1.9$ & $
4.6 \pm 2.3$ & $
3.2 \pm 0.7$ \\
158 5& Yi Se-tol & $
6.0 \pm 1.1$ & $
7.8 \pm 2.3$ & $
9.4 \pm 0.5$ & $
7.6 \pm 1.5$ \\
159 4& Kobayashi Koichi & $
9.5 \pm 0.9$ & $
2.0 \pm 0.7$ & $
2.8 \pm 0.4$ & $
4.0 \pm 1.6$ \\
160 4& Ma Xiaochun & $
8.2 \pm 1.9$ & $
5.2 \pm 1.9$ & $
5.2 \pm 1.8$ & $
6.8 \pm 2.3$ \\
161 4& O Meien & $
2.5 \pm 1.1$ & $
8.2 \pm 2.5$ & $
8.0 \pm 1.6$ & $
4.8 \pm 2.2$ \\
162 4& Rui Naiwei & $
5.5 \pm 1.8$ & $
5.5 \pm 0.5$ & $
9.0 \pm 0.7$ & $
4.0 \pm 1.6$ \\
163 4& Yoda Norimoto & $
7.0 \pm 1.9$ & $
3.8 \pm 2.0$ & $
4.0 \pm 1.9$ & $
3.2 \pm 1.1$ \\
164 3& Go Seigen & $
4.7 \pm 2.5$ & $
6.3 \pm 3.9$ & $
8.0 \pm 0.8$ & $
5.3 \pm 0.9$ \\
165 3& Hane Naoki & $
8.0 \pm 0.8$ & $
3.3 \pm 1.2$ & $
4.0 \pm 0.0$ & $
4.0 \pm 1.4$ \\
166 3& Kato Masao & $
3.0 \pm 0.8$ & $
3.7 \pm 1.7$ & $
8.7 \pm 1.2$ & $
5.7 \pm 2.4$ \\
167 3& Luo Xihe & $
7.3 \pm 0.9$ & $
7.3 \pm 2.5$ & $
7.7 \pm 0.9$ & $
6.0 \pm 1.4$ \\
168 3& Yamashita Keigo & $
2.0 \pm 0.0$ & $
7.3 \pm 2.5$ & $
9.3 \pm 0.5$ & $
4.0 \pm 1.6$ \\
169 2& Chen Yaoye & $
6.0 \pm 1.0$ & $
4.0 \pm 1.0$ & $
6.0 \pm 1.0$ & $
5.5 \pm 0.5$ \\
170 2& Fujisawa Hideyuki & $
3.5 \pm 0.5$ & $
9.0 \pm 1.0$ & $
7.0 \pm 0.0$ & $
4.0 \pm 0.0$ \\
171 2& Honinbo Shusaku & $
8.5 \pm 0.5$ & $
2.0 \pm 1.0$ & $
4.5 \pm 2.5$ & $
4.0 \pm 2.0$ \\
172 2& Miyazawa Goro & $
1.5 \pm 0.5$ & $
10.0 \pm 0.0$ & $
9.5 \pm 0.5$ & $
4.0 \pm 1.0$ \\
173 2& Takao Shinji & $
5.0 \pm 1.0$ & $
3.5 \pm 0.5$ & $
5.5 \pm 1.5$ & $
4.5 \pm 0.5$ \\
174 2& Yuki Satoshi & $
3.0 \pm 1.0$ & $
8.5 \pm 0.5$ & $
9.0 \pm 1.0$ & $
4.5 \pm 0.5$ \\
178 \caption[Expert-based style aspects of selected professionals
]{
179 Expert-based evaluation of styles of selected Professionals, including
180 standard deviation of their answers.
181 Only the players that were evaluated by two or more experts
182 are included. Number of experts who evaluated the particular player is given in first column.
}
187 \section{Acknowledgement
}
188 We would like to thank the following people for filling the questionnaire (in the order
189 of submitted answers):
191 Motoki Noguchi
7-dan, Alexander Dinerchtein
3-pro, Vít Brunner
4-dan,
192 Vladimír Daněk
5-dan and Lukáš Podpěra
5-dan.
195 \bibliographystyle{abbrvnat
}
196 \bibliography{style_consensus
}