Presentation 2018-07-07
Masayuki Irie, Hajime Sasaki, Ichiro Sakata,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) When evaluating restaurants, people consider various aspects. For example, they include taste, quality of service, atmosphere, price. Only showing a unified score makes it difficult to recognize which aspects of restaurants are assessed. The purpose of this research is to identify resembling and competing restaurants by comparing latent topics discovered from reviews by useing Latent Dirichlet Allocation (LDA) algorithm, and recommend similar ones to restaurant management - the users of the system. In this study the review data in Yelp Dataset, which is an open dataset that provides, is analyzed to find latent subtopics. The restaurants which have the same topics are regarded as competitors. In addition, focusing on other features, such as review score, encourages the discovery of more similarities and differences. This paper will be useful to assist restaurants in improving their service.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Yelp / review / aspect / competitor / reccomendation / LDA / service improvement
Paper # NLC2018-8
Date of Issue 2018-06-29 (NLC)

Conference Information
Committee NLC / IPSJ-ICS
Conference Date 2018/7/6(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Hokkaido University
Topics (in Japanese) (See Japanese page)
Topics (in English) Application of natural language processing and intelligent systems, and general topic of NLP
Chair Takeshi Sakaki(Hottolink)
Vice Chair Mitsuo Yoshida(Toyohashi Univ. of Tech.) / Kazutaka Shimada(Kyushu Inst. of Tech.)
Secretary Mitsuo Yoshida(Ryukoku Univ.) / Kazutaka Shimada(NTT)
Assistant Takeshi Kobayakawa(NHK) / Hiroki Sakaji(Univ. of Tokyo)

Paper Information
Registration To Technical Committee on Natural Language Understanding and Models of Communication / Special Interest Group on Intelligence and Complex Systems
Language JPN-ONLY
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English)
Sub Title (in English)
Keyword(1) Yelp
Keyword(2) review
Keyword(3) aspect
Keyword(4) competitor
Keyword(5) reccomendation
Keyword(6) LDA
Keyword(7) service improvement
1st Author's Name Masayuki Irie
1st Author's Affiliation The University of Tokyo(UTokyo)
2nd Author's Name Hajime Sasaki
2nd Author's Affiliation The University of Tokyo(UTokyo)
3rd Author's Name Ichiro Sakata
3rd Author's Affiliation The University of Tokyo(UTokyo)
Date 2018-07-07
Paper # NLC2018-8
Volume (vol) vol.118
Number (no) NLC-122
Page pp.pp.75-80(NLC),
#Pages 6
Date of Issue 2018-06-29 (NLC)