Presentation 2022-07-04
Developing a Part-Oriented Aspect Term Extractor for Part-Specific Review Classification
Shogo Anda, Masato Kikuchi, Tadachika Ozono,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Aspect term extraction is the extraction of terms representing entities and attributes in the target domain of sentiment analysis. We aim to realize sentiment analysis that considers attributes according to part types in product reviews for review classification by product parts. Therefore, we are developing a part-oriented aspect term extraction method to consider the part categories. This paper describes a part-oriented aspect term extraction method using GPT-3 and BERTopic. Our experiments showed that our method could extract 66% of the aspect terms, and topic modeling with text embedding by Sentence-BERT could discover 71% of the attribute type set for each part type.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Aspect Category Extraction / Aspect Analysis / Review Classification / GPT-3 / BERTopic
Paper # AI2022-13
Date of Issue 2022-06-27 (AI)

Conference Information
Committee AI
Conference Date 2022/7/4(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Yuichi Sei(Univ. of Electro-Comm.)
Vice Chair Yuko Sakurai(AIST) / Tadachika Ozono(Nagoya Inst. of Tech.)
Secretary Yuko Sakurai(Tokyo Univ. of Agriculture and Technology) / Tadachika Ozono(Toho Univ.)
Assistant Kazutaka Matsuzaki(Chuo Univ.)

Paper Information
Registration To Technical Committee on Artificial Intelligence and Knowledge-Based Processing
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Developing a Part-Oriented Aspect Term Extractor for Part-Specific Review Classification
Sub Title (in English)
Keyword(1) Aspect Category Extraction
Keyword(2) Aspect Analysis
Keyword(3) Review Classification
Keyword(4) GPT-3
Keyword(5) BERTopic
1st Author's Name Shogo Anda
1st Author's Affiliation Nagoya Institute of Technology(NIT)
2nd Author's Name Masato Kikuchi
2nd Author's Affiliation Nagoya Institute of Technology(NIT)
3rd Author's Name Tadachika Ozono
3rd Author's Affiliation Nagoya Institute of Technology(NIT)
Date 2022-07-04
Paper # AI2022-13
Volume (vol) vol.122
Number (no) AI-94
Page pp.pp.66-71(AI),
#Pages 6
Date of Issue 2022-06-27 (AI)