Generating domain ontology from Chinese customer reviews to analysis fine-gained product quality risk

Shou Lin, Jun Han, Kuldeep Kumar, Jiping Wang

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

With the rapid development of E-commerce in China, quality of the products on online shopping platforms has caused wide concern. Customer reviews, which commented by people who bought the very product, now have been one of the most important resources for analyzing product’s quality risk. We can get fine-gained, aspect-oriented risk information of a product by mining its reviews. Unfortunately, people tend to write reviews with casual grammar or just omit parts of components of a sentence. Both these features will cause negative impacts when parsing the raw customer reviews directly. Thus a knowledge base which is built totally beyond the reviews could be used to analyze it despite the drawbacks above. In this paper, we generate a domain ontology from raw text in the online encyclopedia. It can be viewed as a graph whose nodes represent domain concepts and edges represent the relations between these concepts. In our work, we integrate syntactic tree structure in linear-chain CRFs for recognizing domain concepts and train SVMs and MaxEnt models on elaborate features for clarifying three types of relationship, namely “Attribute-of”, “Part-of” and “Instance-of”. Once the ontology has been built, product properties with potential risk will be extracted by our matching method. Experiment show that our approach achieves 64.4% precision and 82.4% recall on risky property extraction task.

Original languageEnglish
Title of host publicationICCDE 2018 Proceedings of the 2018 International Conference on Computing and Data Engineering
Place of PublicationNew York
PublisherAssociation for Computing Machinery (ACM)
Pages73-77
Number of pages5
VolumePart F137704
ISBN (Print)9781450363938
DOIs
Publication statusPublished - 4 May 2018
Event2018 International Conference on Computing and Data Engineering, ICCDE 2018 - Shanghai, China
Duration: 4 May 20186 May 2018

Conference

Conference2018 International Conference on Computing and Data Engineering, ICCDE 2018
CountryChina
CityShanghai
Period4/05/186/05/18

Fingerprint

Ontology
Electronic commerce
Syntactics
Experiments

Cite this

Lin, S., Han, J., Kumar, K., & Wang, J. (2018). Generating domain ontology from Chinese customer reviews to analysis fine-gained product quality risk. In ICCDE 2018 Proceedings of the 2018 International Conference on Computing and Data Engineering (Vol. Part F137704, pp. 73-77). New York: Association for Computing Machinery (ACM). https://doi.org/10.1145/3219788.3219797, https://doi.org/10.1145/3219788.3219797
Lin, Shou ; Han, Jun ; Kumar, Kuldeep ; Wang, Jiping. / Generating domain ontology from Chinese customer reviews to analysis fine-gained product quality risk. ICCDE 2018 Proceedings of the 2018 International Conference on Computing and Data Engineering. Vol. Part F137704 New York : Association for Computing Machinery (ACM), 2018. pp. 73-77
@inproceedings{e99607a9018a4639828ffe7c8ff13632,
title = "Generating domain ontology from Chinese customer reviews to analysis fine-gained product quality risk",
abstract = "With the rapid development of E-commerce in China, quality of the products on online shopping platforms has caused wide concern. Customer reviews, which commented by people who bought the very product, now have been one of the most important resources for analyzing product’s quality risk. We can get fine-gained, aspect-oriented risk information of a product by mining its reviews. Unfortunately, people tend to write reviews with casual grammar or just omit parts of components of a sentence. Both these features will cause negative impacts when parsing the raw customer reviews directly. Thus a knowledge base which is built totally beyond the reviews could be used to analyze it despite the drawbacks above. In this paper, we generate a domain ontology from raw text in the online encyclopedia. It can be viewed as a graph whose nodes represent domain concepts and edges represent the relations between these concepts. In our work, we integrate syntactic tree structure in linear-chain CRFs for recognizing domain concepts and train SVMs and MaxEnt models on elaborate features for clarifying three types of relationship, namely “Attribute-of”, “Part-of” and “Instance-of”. Once the ontology has been built, product properties with potential risk will be extracted by our matching method. Experiment show that our approach achieves 64.4{\%} precision and 82.4{\%} recall on risky property extraction task.",
author = "Shou Lin and Jun Han and Kuldeep Kumar and Jiping Wang",
year = "2018",
month = "5",
day = "4",
doi = "10.1145/3219788.3219797",
language = "English",
isbn = "9781450363938",
volume = "Part F137704",
pages = "73--77",
booktitle = "ICCDE 2018 Proceedings of the 2018 International Conference on Computing and Data Engineering",
publisher = "Association for Computing Machinery (ACM)",
address = "United States",

}

Lin, S, Han, J, Kumar, K & Wang, J 2018, Generating domain ontology from Chinese customer reviews to analysis fine-gained product quality risk. in ICCDE 2018 Proceedings of the 2018 International Conference on Computing and Data Engineering. vol. Part F137704, Association for Computing Machinery (ACM), New York, pp. 73-77, 2018 International Conference on Computing and Data Engineering, ICCDE 2018, Shanghai, China, 4/05/18. https://doi.org/10.1145/3219788.3219797, https://doi.org/10.1145/3219788.3219797

Generating domain ontology from Chinese customer reviews to analysis fine-gained product quality risk. / Lin, Shou; Han, Jun; Kumar, Kuldeep; Wang, Jiping.

ICCDE 2018 Proceedings of the 2018 International Conference on Computing and Data Engineering. Vol. Part F137704 New York : Association for Computing Machinery (ACM), 2018. p. 73-77.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

TY - GEN

T1 - Generating domain ontology from Chinese customer reviews to analysis fine-gained product quality risk

AU - Lin, Shou

AU - Han, Jun

AU - Kumar, Kuldeep

AU - Wang, Jiping

PY - 2018/5/4

Y1 - 2018/5/4

N2 - With the rapid development of E-commerce in China, quality of the products on online shopping platforms has caused wide concern. Customer reviews, which commented by people who bought the very product, now have been one of the most important resources for analyzing product’s quality risk. We can get fine-gained, aspect-oriented risk information of a product by mining its reviews. Unfortunately, people tend to write reviews with casual grammar or just omit parts of components of a sentence. Both these features will cause negative impacts when parsing the raw customer reviews directly. Thus a knowledge base which is built totally beyond the reviews could be used to analyze it despite the drawbacks above. In this paper, we generate a domain ontology from raw text in the online encyclopedia. It can be viewed as a graph whose nodes represent domain concepts and edges represent the relations between these concepts. In our work, we integrate syntactic tree structure in linear-chain CRFs for recognizing domain concepts and train SVMs and MaxEnt models on elaborate features for clarifying three types of relationship, namely “Attribute-of”, “Part-of” and “Instance-of”. Once the ontology has been built, product properties with potential risk will be extracted by our matching method. Experiment show that our approach achieves 64.4% precision and 82.4% recall on risky property extraction task.

AB - With the rapid development of E-commerce in China, quality of the products on online shopping platforms has caused wide concern. Customer reviews, which commented by people who bought the very product, now have been one of the most important resources for analyzing product’s quality risk. We can get fine-gained, aspect-oriented risk information of a product by mining its reviews. Unfortunately, people tend to write reviews with casual grammar or just omit parts of components of a sentence. Both these features will cause negative impacts when parsing the raw customer reviews directly. Thus a knowledge base which is built totally beyond the reviews could be used to analyze it despite the drawbacks above. In this paper, we generate a domain ontology from raw text in the online encyclopedia. It can be viewed as a graph whose nodes represent domain concepts and edges represent the relations between these concepts. In our work, we integrate syntactic tree structure in linear-chain CRFs for recognizing domain concepts and train SVMs and MaxEnt models on elaborate features for clarifying three types of relationship, namely “Attribute-of”, “Part-of” and “Instance-of”. Once the ontology has been built, product properties with potential risk will be extracted by our matching method. Experiment show that our approach achieves 64.4% precision and 82.4% recall on risky property extraction task.

UR - http://www.scopus.com/inward/record.url?scp=85053707791&partnerID=8YFLogxK

U2 - 10.1145/3219788.3219797

DO - 10.1145/3219788.3219797

M3 - Conference contribution

SN - 9781450363938

VL - Part F137704

SP - 73

EP - 77

BT - ICCDE 2018 Proceedings of the 2018 International Conference on Computing and Data Engineering

PB - Association for Computing Machinery (ACM)

CY - New York

ER -

Lin S, Han J, Kumar K, Wang J. Generating domain ontology from Chinese customer reviews to analysis fine-gained product quality risk. In ICCDE 2018 Proceedings of the 2018 International Conference on Computing and Data Engineering. Vol. Part F137704. New York: Association for Computing Machinery (ACM). 2018. p. 73-77 https://doi.org/10.1145/3219788.3219797, https://doi.org/10.1145/3219788.3219797