Paper Abstract and Keywords |
Presentation |
2010-11-20 08:30
ZeoBro: a Personal Life Assistance Service Platform on a PC-Cluster Yuta Tajima, Takefumi Miyoshi (UEC), Sayaka Akioka (Waseda Univ.), Tatsuya Goto, Hidetsugu Irie, Tsutomu Yoshinaga (UEC) IN2010-93 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
Most of personal life assistance services use activity data such as GPS location. Most of them require user histories before starting the services. We propose ZeoBro which recommends spot data parsed from blogs written by the other user. ZeoBro uses collaborative filtering in order to make spot relations. However, the processing time increases if volume of blog data and spot data increase. To solve the problem, we implemented this system on PC-cluster and processing was made parallel. We report our initial evaluation for ZeoBro based on its analysis for 800,000 blog pages. We also show the computation time reduction with varying the number of nodes in the PC-cluster. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
Personal Life Assistance Service / Geomedia / Document Analysis / Distributed Computing / / / / |
Reference Info. |
IEICE Tech. Rep., vol. 110, no. 289, IN2010-93, pp. 71-76, Nov. 2010. |
Paper # |
IN2010-93 |
Date of Issue |
2010-11-11 (IN) |
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 |
IN2010-93 |
|