Presentation 1995/10/16
Evaluation of a Natural Language Interface for Image Retrieval
Shouji Harada, Hiroyuki Tanaka, Hiromasa Nakatani, Yukihiro Itoh,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We have been developing an image data retrieval system that can accept inquiries in natural language. In this paper, We evaluate the performance of a natural language interface of the system. To compare NLI (natural language interface) with GUI (graphical user interface), we make 32 subjects use the system and measure success rate for the retrieval, number of interactions and thinking time. We tell the students to retrieve some pictures of chairs by using each system, NLI and GUI. In all the cases, almost all the subjects succeed to retrieve a chair that satisfies them. To retrieve the preference, they input 3 or 4 sentences on average. As a result, we confirm that our natural language interface is as effective as GUI generally and some users prefer natural language interface.
Keyword(in Japanese) (See Japanese page)
Keyword(in English)
Paper # IE95-63
Date of Issue

Conference Information
Committee IE
Conference Date 1995/10/16(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 Image Engineering (IE)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Evaluation of a Natural Language Interface for Image Retrieval
Sub Title (in English)
Keyword(1)
1st Author's Name Shouji Harada
1st Author's Affiliation Department of Computer Science, Shizuoka University()
2nd Author's Name Hiroyuki Tanaka
2nd Author's Affiliation Department of Computer Science, Shizuoka University
3rd Author's Name Hiromasa Nakatani
3rd Author's Affiliation Department of Computer Science, Shizuoka University
4th Author's Name Yukihiro Itoh
4th Author's Affiliation Department of Computer Science, Shizuoka University
Date 1995/10/16
Paper # IE95-63
Volume (vol) vol.95
Number (no) 292
Page pp.pp.-
#Pages 8
Date of Issue