1998 Fiscal Year Final Research Report Summary
Study of Collecting User Operation Log in Ordinary Environments and Comparison with Controlled Experiment Data
| Project/Area Number |
09480053
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
計算機科学
|
|---|
| Research Institution | Osaka University |
|---|
Principal Investigator |
TOKURA Nobuki Graduate School of Engineering Science, Professor, 大学院・基礎工学研究科, 教授 (10029423)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
YASUTOME Seigo Southern Osaka University, Associate Professor, 助教授 (50252721)
MASUDA Hideo Education Center for Information Processing, Research Associate, 情報処理教育センター, 助手 (90304063)
NISHIDA Tomohiro Education Center for Information Processing, Research Associate, 情報処理教育センター, 助手 (00283820)
SAITOH Akinori Education Center for Information Processing, Assistant Professor, 情報処理教育センター, 講師 (20235021)
TSUJINO Yoshihiro Graduate School of Engineering Science, Associate Professor, 大学院・基礎工学研究科, 助教授 (20172009)
|
|---|
| Project Period (FY) |
1997 – 1998
|
| Keywords | Ordinary Operation Log / Controlled Experiment / User Analysis / Target Selection / Mouse / Pointing Error |
|---|
| Research Abstract |
In order to characterize the influences of observed conditions, such as controlled experiments, we develop the way to collect user GUI operation log in ordinary environment, such as students' computer room. We gather 161 students' operation log for S months. By comparing about the basic GUI operation time, we obtain the results shown below.
・ In controlled experiment, objects tend to move mice by their wrists and upper arms and stop their operations as fast as possible when they select large targets, such as a window.
・ The operation time in the controlled experiment is shorter because objects concentrate on their tasks. At the students' computer room, the distribution of key-stroke intervals between students is small and characteristic.
Furthermore, we analyze the ordinary environment data to inquire the cause of mouse pointing errors. We find two types of error, one is an unintentional mouse motion affected by the side effect of a button operation and the other is a synchronization error between a button operation and a move operation. From that analysis, we make an algorithm to guess the coordinates that the user want to point and confirm to reduce pointing errors by using our algorithm.
|
