The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Lorsque les individus effectuent un achat auprès de sources en ligne, ils peuvent manquer de connaissances directes sur le produit. Dans de tels cas, ils jugeront la qualité de l’article à l’aide des avis publiés par d’autres consommateurs. Il est donc important de déterminer si les commentaires sur un produit sont crédibles. Le plus souvent, les recherches conventionnelles sur la crédibilité des commentaires ont eu recours à des méthodes d’apprentissage automatique supervisé, qui présentent l’inconvénient de nécessiter de grandes quantités de données de formation. Cet article propose une méthode non supervisée pour juger de la crédibilité des commentaires basée sur le modèle Biterm Sentiment Latent Dirichlet Allocation (BS-LDA). En utilisant cette approche, nous avons d'abord dérivé quelques distributions et calculé le score de crédibilité de chaque commentaire via celles-ci. La crédibilité d'un commentaire a été jugée selon qu'il atteignait ou non un score seuil. Nos résultats expérimentaux utilisant les commentaires d'Amazon.com ont démontré que la performance globale de notre approche peut jouer un rôle important dans la détermination de la crédibilité des commentaires dans certaines situations.
Xuan WANG
Shanghai University
Bofeng ZHANG
Shanghai University
Mingqing HUANG
Shanghai University
Furong CHANG
Kashgar University
Zhuocheng ZHOU
Shanghai University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Xuan WANG, Bofeng ZHANG, Mingqing HUANG, Furong CHANG, Zhuocheng ZHOU, "Improved LDA Model for Credibility Evaluation of Online Product Reviews" in IEICE TRANSACTIONS on Information,
vol. E102-D, no. 11, pp. 2148-2158, November 2019, doi: 10.1587/transinf.2018EDP7243.
Abstract: When individuals make a purchase from online sources, they may lack first-hand knowledge of the product. In such cases, they will judge the quality of the item by the reviews other consumers have posted. Therefore, it is significant to determine whether comments about a product are credible. Most often, conventional research on comment credibility has employed supervised machine learning methods, which have the disadvantage of needing large quantities of training data. This paper proposes an unsupervised method for judging comment credibility based on the Biterm Sentiment Latent Dirichlet Allocation (BS-LDA) model. Using this approach, first we derived some distributions and calculated each comment's credibility score via them. A comment's credibility was judged based on whether it achieved a threshold score. Our experimental results using comments from Amazon.com demonstrated that the overall performance of our approach can play an important role in determining the credibility of comments in some situation.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDP7243/_p
Copier
@ARTICLE{e102-d_11_2148,
author={Xuan WANG, Bofeng ZHANG, Mingqing HUANG, Furong CHANG, Zhuocheng ZHOU, },
journal={IEICE TRANSACTIONS on Information},
title={Improved LDA Model for Credibility Evaluation of Online Product Reviews},
year={2019},
volume={E102-D},
number={11},
pages={2148-2158},
abstract={When individuals make a purchase from online sources, they may lack first-hand knowledge of the product. In such cases, they will judge the quality of the item by the reviews other consumers have posted. Therefore, it is significant to determine whether comments about a product are credible. Most often, conventional research on comment credibility has employed supervised machine learning methods, which have the disadvantage of needing large quantities of training data. This paper proposes an unsupervised method for judging comment credibility based on the Biterm Sentiment Latent Dirichlet Allocation (BS-LDA) model. Using this approach, first we derived some distributions and calculated each comment's credibility score via them. A comment's credibility was judged based on whether it achieved a threshold score. Our experimental results using comments from Amazon.com demonstrated that the overall performance of our approach can play an important role in determining the credibility of comments in some situation.},
keywords={},
doi={10.1587/transinf.2018EDP7243},
ISSN={1745-1361},
month={November},}
Copier
TY - JOUR
TI - Improved LDA Model for Credibility Evaluation of Online Product Reviews
T2 - IEICE TRANSACTIONS on Information
SP - 2148
EP - 2158
AU - Xuan WANG
AU - Bofeng ZHANG
AU - Mingqing HUANG
AU - Furong CHANG
AU - Zhuocheng ZHOU
PY - 2019
DO - 10.1587/transinf.2018EDP7243
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 2019
AB - When individuals make a purchase from online sources, they may lack first-hand knowledge of the product. In such cases, they will judge the quality of the item by the reviews other consumers have posted. Therefore, it is significant to determine whether comments about a product are credible. Most often, conventional research on comment credibility has employed supervised machine learning methods, which have the disadvantage of needing large quantities of training data. This paper proposes an unsupervised method for judging comment credibility based on the Biterm Sentiment Latent Dirichlet Allocation (BS-LDA) model. Using this approach, first we derived some distributions and calculated each comment's credibility score via them. A comment's credibility was judged based on whether it achieved a threshold score. Our experimental results using comments from Amazon.com demonstrated that the overall performance of our approach can play an important role in determining the credibility of comments in some situation.
ER -