Presentation 2016-03-03
Automatic Extraction of Information for "Palatability" in Microblogs: A Proposal of A Restaurant Retrieval System based on Personal Preferences and Real-time Data
Maki Morita, Eiji Aramaki, Akiyo Nadamoto, Mai Miyabe,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The restaurant information searches on Internet are widely available now, but such searches have problems of providing users exact information that have been looked for, and even sometimes the information are not up-to-date. In this study, we propose a restaurant retrieval system based on personal preferences for providing up-to-date information by using the expressions in microblog posts that contain sizzle feelings towards food and beverages, such as ``MOCHI-MOCHI" and ``SAKU-SAKU." In order to develop such system, we have constructed the automatic classifier for tweets that contain sizzle related words. We have tried classifying 5 sizzle related words, and we have obtained the following 3 findings:1) Our classifier is capable of classifying sizzle-related tweets with high accuracy. 2) Classification accuracy can be improved by using context information as a feature. 3) Sizzle related tweets can be classified with high accuracy as well as low accuracy depends on the sizzle related word that construct the classifier.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Sizzle related words / Microblog / Twitter / Automatic classification / Restaurant recommendation
Paper # LOIS2015-74
Date of Issue 2016-02-25 (LOIS)

Conference Information
Committee LOIS
Conference Date 2016/3/3(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Central Community Center, Miyakojima-City
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Manabu Okamoto(NTT)
Vice Chair Hiroyuki Nishi(Sojo Univ.)
Secretary Hiroyuki Nishi(Tsuda College)
Assistant Yu Ichifuji(NII)

Paper Information
Registration To Technical Committee on Life Intelligence and Office Information Systems
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Automatic Extraction of Information for "Palatability" in Microblogs: A Proposal of A Restaurant Retrieval System based on Personal Preferences and Real-time Data
Sub Title (in English)
Keyword(1) Sizzle related words
Keyword(2) Microblog
Keyword(3) Twitter
Keyword(4) Automatic classification
Keyword(5) Restaurant recommendation
1st Author's Name Maki Morita
1st Author's Affiliation Wakayama University(Wakayama Univ.)
2nd Author's Name Eiji Aramaki
2nd Author's Affiliation Nara Institute of Science and Technology(NAIST)
3rd Author's Name Akiyo Nadamoto
3rd Author's Affiliation Konan University(Konan Univ.)
4th Author's Name Mai Miyabe
4th Author's Affiliation Wakayama University(Wakayama Univ.)
Date 2016-03-03
Paper # LOIS2015-74
Volume (vol) vol.115
Number (no) LOIS-486
Page pp.pp.63-68(LOIS),
#Pages 6
Date of Issue 2016-02-25 (LOIS)