Presentation 1997/1/25
An Analysis for the Neural Network Supporting Users' selection of Menu Items
Hisayoshi ITO, Hiroki ITO, Kanji AKAHORI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This study is to support users with the neural network by reasoning correct tasks from their wrong selected menu items. In this report, we analyze the mechanism of the neural network's reasoning for supporting users. We investigated the neural network whose input is wrong selection of menu items and whose output is the tasks which is seemed to e the correct one corresponding the input. In the analysis we focused on weights between hidden layor and outputs, and found each cell's tendency what type of tasks in the outputs it is likely to make active or inactive, and also had meaning of the factors. And we try to argue that these characteristics are similar to the subtasks made by task analysis.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) neural network / user interface / task analysis / cognitive structure
Paper # ET96-111
Date of Issue

Conference Information
Committee ET
Conference Date 1997/1/25(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 Educational Technology (ET)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) An Analysis for the Neural Network Supporting Users' selection of Menu Items
Sub Title (in English)
Keyword(1) neural network
Keyword(2) user interface
Keyword(3) task analysis
Keyword(4) cognitive structure
1st Author's Name Hisayoshi ITO
1st Author's Affiliation Tokyo Institute of Technology()
2nd Author's Name Hiroki ITO
2nd Author's Affiliation NTT Advanced Technology Co., Inc.
3rd Author's Name Kanji AKAHORI
3rd Author's Affiliation Tokyo Institute of Technology
Date 1997/1/25
Paper # ET96-111
Volume (vol) vol.96
Number (no) 500
Page pp.pp.-
#Pages 8
Date of Issue