Details of a Researcher

ITO Yudai

写真a

Affiliation	Faculty of Informatics and Data Science Department of Informatics and Data Science
Date of Birth	1997
Homepage URL	https://adeos6.ie.akita-u.ac.jp/
Mail Address

To the head of this page.▲

Research Interests 【 display / non-display 】

Machine learning
Eye tracking

To the head of this page.▲

Graduating School 【 display / non-display 】

2016.04

-

2020.03

Akita University Faculty of Science and Engineering Department of Mathematical Science and Electrical-Electronic-Computer Engineering Human-Centered Computing Course Graduated

To the head of this page.▲

Graduate School 【 display / non-display 】

2022.04

-

2025.03

Akita University Graduate School, Division of Science and Engineering Department of Integrated Engineering Science Field of Mathematical Science and Electrical-Electronic-Computer Engineering Doctor's Degree Program Completed
2020.04

-

2022.03

Akita University Graduate School, Division of Science and Engineering Department of Mathematical Science and Electrical-Electronic-Computer Engineering Human-Centered Computing Course Master's Degree Program Completed

To the head of this page.▲

Degree 【 display / non-display 】

Akita University - Master of Engineering
Akita University - Bachelor of Engineering

To the head of this page.▲

Campus Career 【 display / non-display 】

2025.04

-

Now

Akita University Faculty of Informatics and Data Science Department of Informatics and Data Science Assistant Professor

To the head of this page.▲

Academic Society Affiliations 【 display / non-display 】

2022.03

-

Now

Japan

The Institute of Electrical Engineers of Japan
2020.06

-

Now

Japan

Information Processing Society of Japan

To the head of this page.▲

Research Areas 【 display / non-display 】

Informatics / Learning support system
Informatics / Human interface and interaction
Informatics / Kansei informatics

To the head of this page.▲

Qualification acquired 【 display / non-display 】

Fundamental Information Engineer/Information Processing Engineer, Class 2

To the head of this page.▲

Thesis for a degree 【 display / non-display 】

A method for correcting gaze coordinate errors measured and setting a fixation threshold to detect difficult-to-read words during silent reading of Japanese text

Yudai Ito

2025.03

Single author

This study introduces a novel approach to improve the detection of difficult-to-read words in Japanese text using eye-tracking technology. There are two problems to detect the difficult-to-read word.
First problem is position error of eye-tracking data. To solve the problem, existing methods are focus only on y-axis correction. However, the proposed method is x-axis correction by projective transformation. In experimental evaluation, proposed method has demonstrated that it improves the detection rate of difficult-to-read words when combined with existing methods.
Second problem is determination method of fixation duration thresholds for a critical parameter for identifying difficult-to-read word. Unlike standard approaches that rely on fixed 50 ms or 100 ms steps, this research proposes a more efficient threshold search with larger step intervals, such as 200 ms, while considering individual differences in reading behavior. This method reduces computational costs and enhances the usability of detection systems. The experimental results validate the efficacy of the proposed threshold determination method, which outperforms the detection accuracy of difficult-to-read word than previous thresholds, including those fixed at 1,000 ms or 800 ms.
These results suggest that integrating adaptive thresholds and accurate x and y-axis correction provides a more reliable mechanism for identifying difficult-to-read portions of horizontal Japanese text. Practical applications of this research include reviewing text readability in educational and professional documents, supporting language learners by providing targeted feedback such as automated list up of difficult-to-read words and supplementary explanation.

To the head of this page.▲

Research Achievements 【 display / non-display 】

◆Original paper【 display / non-display 】

Automated Correction of Eye-Tracking Data for Detecting Difficult-to-Read Characters in Japanese Text

Yudai Ito, Chikako Ishizawa, Yoichi Kageyama

IEEJ Transactions on Electrical and Electronic Engineering ( IEEJ Transactions on Electrical and Electronic Engineering ) 19 ( 6 ) 975 - 984 2024.06 [Refereed]

Research paper (journal) Domestic Co-author

DOI

Threshold Determination Method for Detecting the Gazing Areas on a PC Screen While Silent Reading Japanese Texts

Yudai Ito, Chikako Ishizawa, Yoichi Kageyama

IEEJ Transactions on Electrical and Electronic Engineering ( IEEJ Transactions on Electrical and Electronic Engineering ) 18 ( 5 ) 714 - 721 2023.05 [Refereed]

Research paper (journal) Domestic Co-author

DOI

Method for Detecting Sources Cited with Browsing to Prevent Reference Omissions

C. Ishizawa, M. Hiroyuki, Y. Ito, Y. Kageyama

IEEJ Transactions on Electrical and Electronic Engineering ( IEEJ Transactions on Electrical and Electronic Engineering ) 18 ( 1 ) 150 - 152 2023.01 [Refereed]

Research paper (journal) Domestic Co-author

DOI

To the head of this page.▲

◆International conference proceedings【 display / non-display 】

Facial Video Analysis using Eye-tracking data during Japanese Text Reading

Y. Ito, C. Ishizawa, Y. Kageyama

2024 Joint 13th International Conference on Soft Computing and Intelligent Systems and 25th International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2024 ( 2024 Joint 13th International Conference on Soft Computing and Intelligent Systems and 25th International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2024 ) 1 - 2 2024.11 [Refereed]