2009 Sixth International Conference on Fuzzy Systems and Knowledge Discovery 2009
DOI: 10.1109/fskd.2009.200
|View full text |Cite
|
Sign up to set email alerts
|

Opinion Analysis of Product Reviews

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
4
0

Year Published

2010
2010
2022
2022

Publication Types

Select...
2
2
2

Relationship

0
6

Authors

Journals

citations
Cited by 8 publications
(4 citation statements)
references
References 8 publications
0
4
0
Order By: Relevance
“…From the perspective of supervised methods, Jin et al (2009) used the hidden Markov model and bootstrapping algorithm to extract camera attributes from online reviews; they argued that their proposed strategy could recognize new features without relying on natural language processing technologies. Zhang et al (2009) and Xu et al (2011) used conditional random fields, domain dictionary and language features to extract product aspects, but it takes more time to construct and update the domain dictionary. Moreover, topic models (Andrzejewski et al, 2009;Xu et al, 2015) and real-time review streams (Dragoni et al, 2019) have also been used in attributes extraction.…”
Section: Attribute Extractionmentioning
confidence: 99%
“…From the perspective of supervised methods, Jin et al (2009) used the hidden Markov model and bootstrapping algorithm to extract camera attributes from online reviews; they argued that their proposed strategy could recognize new features without relying on natural language processing technologies. Zhang et al (2009) and Xu et al (2011) used conditional random fields, domain dictionary and language features to extract product aspects, but it takes more time to construct and update the domain dictionary. Moreover, topic models (Andrzejewski et al, 2009;Xu et al, 2015) and real-time review streams (Dragoni et al, 2019) have also been used in attributes extraction.…”
Section: Attribute Extractionmentioning
confidence: 99%
“…In a machine learning-based attribute extraction method, Zhang et al [13] introduced word-level features in the CRF model and used domain dictionary knowledge as an aid for product attribute extraction. Xu et al [14] introduced shallow syntactic information and heuristic location information and input them to CRF as features, which effectively improved the attribute extraction performance of the model.…”
Section: Related Workmentioning
confidence: 99%
“…The loss value calculation formula is shown in Equation (13), where loss start and loss end represent the loss values of entity head recognition and entity tail recognition, respectively, and loss attribute represents the loss value generated by the attribute sequence labelling. α, β, γ ∈ ½0, 1 are hyperparameters that control the weighted summation of the threeloss values.…”
Section: Loss Value Calculationmentioning
confidence: 99%
“…Liu [6][7] identifies product features using association mining firstly, and then adjective word nearest to product features is regarded as opinion word. Zhang [8] adopt the CRFs model to find product features with the domain knowledge. Our research is closely related with [3][6] [8], but it differs form others in that we use chunk features and heuristic information to identify product features for CRFs.…”
Section: Related Workmentioning
confidence: 99%