Presentation 1993/9/20
Iconic Selection based on Stochastic Process
Satoru Morita, Toru Ootuka, Minoru Tanaka,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We have developed an iconic work space,in which the tasks of overlaping an icon on another icon simulate the work in real world. End-users grasp the work space by observing icon sets.When a great number of icons exists,icons,not relating to the users′ work,preve nt them from identifying for a target icon.To solve this problem hierachical environment of icons was proposed.However the hierachy proposed by the system designer sometimes doesn′t match with the e nd-users need.So our system selects icons using the task history without hierachical relation of icons defined by the system designer.We formulated this problem as time sequence estimation problem using stochastic process.The effectiveness of the proposed model on the batch dependent and class dependent processing were evaluated.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) User Interface / Stochastic Process / Iconic Programming system
Paper # AI93-45
Date of Issue

Conference Information
Committee AI
Conference Date 1993/9/20(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Artificial Intelligence and Knowledge-Based Processing (AI)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Iconic Selection based on Stochastic Process
Sub Title (in English)
Keyword(1) User Interface
Keyword(2) Stochastic Process
Keyword(3) Iconic Programming system
1st Author's Name Satoru Morita
1st Author's Affiliation Yamaguchi University()
2nd Author's Name Toru Ootuka
2nd Author's Affiliation Yamaguchi University
3rd Author's Name Minoru Tanaka
3rd Author's Affiliation Yamaguchi University
Date 1993/9/20
Paper # AI93-45
Volume (vol) vol.93
Number (no) 240
Page pp.pp.-
#Pages 8
Date of Issue