| IEICE Technical Committee Submission System Conference Paper's Information |
Online Proceedings [Sign in] Tech. Rep. Archives |
| Go Top Page | Go Previous | [Japanese] / [English] |
| Paper Abstract and Keywords | ||
| Presentation | 2017-09-01 14:35 Shiori Wada, Shigetaka Suwa, Takeshi Wakumura, Toru Sugimoto, Midori Sugaya (SIT)  CNR2017-12 |
|
| Abstract | (in Japanese) | (See Japanese page) |
| (in English) | [Introduction] There is lost child search at a shopping mall as one of the applicability of the service by more than one robots. A robot collects information from a person and analyzes, and lost child search has to share. Locomotive robot communicates using ad hoc network and each robot is to share lost child information interactively, and this research proposes the system that a lost child can be searched. In this way, there are two merits. One is that it makes a child possess no devices beforehand. The other is that it isn't necessary to register child's photograph of the face with which beforehand. [Way] I undertake information gathering by a natural language from the user by the dialog system loaded into a robot. The user supposes a parent of lost child, surrounding adults who saw a lost child, and lost child. User talk with one robot which is nearest from he or she. Parent of lost child utters child's first name, his or her first name, and feature of lost child's appearance. Surrounding adults who saw a lost child utters feature of the child's appearance. Lost child utters his or her first name and his or her person's first name. If a robot collects information from the user, the information collected from the user is shared with other robots by communicating by ad hoc network. When the name of the parent and the name of the lost child a parent and a lost child input are identical respectively, and cosine similarity between feature of lost child which input by his or her parent and feature of lost child which input by surrounding adults is more than 0.6, I consider I succeeded in lost child's search. [Result] In evaluation experiment, a specific child showed by animation and picture to a subject. In this experiment, I considered the specific child to be a lost child. The feature of the lost child lost child's parent would input was input beforehand by me, and all subjects were made setting as the surrounding adult. When the cosine similarity between the feature input by me and the feature input by subjects was more than 0.6, it was made experimental success. Through this experiment, I found three things. First, the experimental success rate was higher when the subjects who have a child than when the subjects who don't have a child. Second, when subjects input gender of the child, cosine similarity was high. Third, subjects input information that is not considered by this system, cosine similarity was low because of shortage of data. [Consideration] The gender of the lost child, worn clothes, the height and a frame were considered as input data by this system. However, in evaluation experiment, there were the subjects who included the approximate age, feature of the face, action he or she did, and his or her atmosphere of the lost child. Consideration of these information is also necessary because the cosine similarity was data shortage and could confirm the low thing when such information was input much from a subject. In addition, I found that if one surrounding adult couldn't remember the feature of lost child perfectly, other surrounding adult remembered the feature that is not remembered by user before or remembered the feature that is remembered by user before, too, cosine similarity was high. Finally, it is necessary to think about a case when parent of lost child doesn't remember the clothes the lost child worn or remember the clothes the lost child worn incorrectly. |
|
| Keyword | (in Japanese) | (See Japanese page) |
| (in English) | lost child / dialog system / natural language / cosine similarity / / / / | |
| Reference Info. | IEICE Tech. Rep., vol. 117, no. 198, CNR2017-12, pp. 19-24, Sept. 2017. | |
| Paper # | CNR2017-12 | |
| Date of Issue | 2017-08-25 (CNR) | |
| ISSN | Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 | |
| Copyright and reproduction |
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) | |
| Download PDF | CNR2017-12 |
|
| Conference Information | |
| Committee | CNR |
| Conference Date | 2017-09-01 - 2017-09-01 |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Paper Information | |
| Registration To | CNR |
| Conference Code | 2017-09-CNR |
| Language | Japanese without English title) |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | |
| Sub Title (in English) | |
| Keyword(1) | lost child |
| Keyword(2) | dialog system |
| Keyword(3) | natural language |
| Keyword(4) | cosine similarity |
| Keyword(5) | |
| Keyword(6) | |
| Keyword(7) | |
| Keyword(8) | |
| 1st Author's Name | Shiori Wada |
| 1st Author's Affiliation | Shibaura Institute of Technology (SIT) |
| 2nd Author's Name | Shigetaka Suwa |
| 2nd Author's Affiliation | Shibaura Institute of Technology (SIT) |
| 3rd Author's Name | Takeshi Wakumura |
| 3rd Author's Affiliation | Shibaura Institute of Technology (SIT) |
| 4th Author's Name | Toru Sugimoto |
| 4th Author's Affiliation | Shibaura Institute of Technology (SIT) |
| 5th Author's Name | Midori Sugaya |
| 5th Author's Affiliation | Shibaura Institute of Technology (SIT) |
| 6th Author's Name | |
| 6th Author's Affiliation | () |
| 7th Author's Name | |
| 7th Author's Affiliation | () |
| 8th Author's Name | |
| 8th Author's Affiliation | () |
| 9th Author's Name | |
| 9th Author's Affiliation | () |
| 10th Author's Name | |
| 10th Author's Affiliation | () |
| 11th Author's Name | |
| 11th Author's Affiliation | () |
| 12th Author's Name | |
| 12th Author's Affiliation | () |
| 13th Author's Name | |
| 13th Author's Affiliation | () |
| 14th Author's Name | |
| 14th Author's Affiliation | () |
| 15th Author's Name | |
| 15th Author's Affiliation | () |
| 16th Author's Name | |
| 16th Author's Affiliation | () |
| 17th Author's Name | |
| 17th Author's Affiliation | () |
| 18th Author's Name | |
| 18th Author's Affiliation | () |
| Speaker | Author-1 |
| Date Time | 2017-09-01 14:35:00 |
| Presentation Time | 20 minutes |
| Registration for | CNR |
| Paper # | CNR2017-12 |
| Volume (vol) | vol.117 |
| Number (no) | no.198 |
| Page | pp.19-24 |
| #Pages | 6 |
| Date of Issue | 2017-08-25 (CNR) |
[Return to Top Page]