「ロボットは東大に入れるか」プロジェクト： 代ゼミセンター模試タスクにおけるエラーの分析

松崎 拓也; 横野 光; 宮尾 祐介; 川添 愛; 狩野 芳伸; 加納 隼人; 佐藤 理史; 東中 竜一郎; 杉山 弘晃; 磯崎 秀樹; 菊井 玄一郎; 堂坂 浩二; 平 博順; 南 泰浩; 新井 紀子

doi:10.5715/jnlp.23.119

Paper

The Todai Robot Project: Error Analysis on the Results of the Yozemi Center Test

Takuya Matsuzaki, Hikaru Yokono, Yusuke Miyao, Ai Kawazoe, Yoshinobu Kano, Hayato Kanou, Satoshi Sato, Ryuichiro Higashinaka, Hiroaki Sugiyama, Hideki Isozaki, Genichiro Kikui, Koji Dosaka, Hirotoshi Taira, Yasuhiro Minami, Noriko H. Arai

Author information

Takuya Matsuzaki
Graduate School of Engineering, Nagoya University
Hikaru Yokono
National Institute of Informatics
Yusuke Miyao
National Institute of Informatics
Ai Kawazoe
National Institute of Informatics
Yoshinobu Kano
Faculty of Informatics, Shizuoka University
Hayato Kanou
Graduate School of Engineering, Nagoya University
Satoshi Sato
Graduate School of Engineering, Nagoya University
Ryuichiro Higashinaka
NTT Communication Science Laboratories, NTT Corporation
Hiroaki Sugiyama
NTT Communication Science Laboratories, NTT Corporation
Hideki Isozaki
Faculty of Computer Science and Systems Engineering, Okayama Prefectural University
Genichiro Kikui
Faculty of Computer Science and Systems Engineering, Okayama Prefectural University
Koji Dosaka
Faculty of Systems Science and Technology, Akita Prefectural University
Hirotoshi Taira
Faculty of Information Science and Technology, Osaka Institute of Technology
Yasuhiro Minami
Graduate School of Information Systems, The University of Electoro-Comunications
Noriko H. Arai
National Institute of Informatics

Keywords: Integrated AI Tasks, University Entrance Examinations, Error Analysis

JOURNAL FREE ACCESS

2016 Volume 23 Issue 1 Pages 119-159

DOI https://doi.org/10.5715/jnlp.23.119

Details

Abstract

The Todai Robot Project aims at integrating various AI technologies including natural language processing (NLP), as well as uncovering novel AI problems that have been missed while the fragmentation of the research field, through the development of software systems that solve university entrance exam problems. Being primarily designed for the measurement of human intellectual abilities, university entrance exam problems serve as an ideal benchmark for AI technologies. They also enable a quantitative comparison between the AI systems and human test takers. This paper analyzes the errors made by the software systems on the mock university entrance exams hosted by a popular preparatory school. Based on the analyses, key problems towards higher system performances and the current issues in the field of NLP are discussed.

Corresponding author

Register with J-STAGE for free!