Information Technology Reference
In-Depth Information
Ta b l e 2
Ranking of the number of pages for the top 50 researchers of The University of
Tokyo
r
∗
r
∗
Name
Name
1: Yasuhiko Arakawa 26: Kazuhiko Saigo
2: Kazunori Kataoka 27: Tadatomo Suga
3: Kohji Kishio 28: Tamio Arai
4: Yuichi Ikuhara 29: Akira Isogai
5: Kazuhiko Ishihara 30: Ryoichi Yamamoto
6: Yasuhiro Iwasawa 31: Takayasu Sakurai
7: Genki Yagawa 32: Michio Yamawaki
8: Kazuhito Hashimoto 33: Hiroshi Harashima
9: Hiroyuki Sakaki
34: Takayoshi Kobayashi
10: Hideki Imai
35: Fumio Tatsuoka
11: Masaharu Oshima
36: Takehiko Kitamori
12: Kazuyuki Aihara
37: Teruyuki Nagamune
13: Kazuro Kikuchi
38: Masahiko Isobe
14: Yoshiaki Nakano
39: Motohiro Kanno
15: Shinichi Uchida
40: Kazuo Hotate
16: Hidenori Takagi
41: Mitsuhiro Shibayama
17: Hiroyuki Fujita
42: Hajime Asama
18: Katsushi Ikeuchi
43: Satoru Tanaka
19: Yutaka Kagawa
44: Isao Shimoyama
20: Nobuo Takeda
45: Yozo Fujino
21: Masaru Miyayama
46: Takayuki Terai
22: Toshiro Higuchi
47: Yoichiro Matsumoto
23: Tsuguo Sawada
48: Nobuhide Kasagi
24: Kiyoharu Aizawa
49: Yoshiyuki Amemiya
25: Kimihiko Hirao
50: Kunihiro Asada
In our experiments, we conducted three-fold cross-validation. In each trial, two
folds of actors are used for training, and one fold for prediction. The results re-
ported in this section are those averaged over three trials. We use Spearman's rank
correlation coefficient (
ρ
) [11] to measure the pairwise ranking correlation.
2
i
6
Σ
ρ
=
1
−
(5)
n
(
n
2
−
1
)
In that equation,
d
i
signifies the difference between the ranks of corresponding
values
X
i
and
Y
i
.
5.2
Ranking Results
First, we rank researchers on different network rankings. Table 3 presents the degree
centrality rankings of different types of networks in researcher networks. Results
show that
Yutaka Kagawa
has good degree centrality on a cooc network of Japanese
