MétaCan
Menu
Retour à la cohorte
Enregistrement W294906217

Helpful or Unhelpful: A Linear Approach for Ranking Product Reviews

2010· article· en· W294906217 sur OpenAlex
Richong Zhang, Thomas Tran

Pourquoi ce travail est dans la base

Une base qui oublie comment elle a trouvé un travail ne peut pas être vérifiée. Voici les voies qui ont admis celui-ci.

affAu moins un auteur déclare une institution canadienne dans l'instantané OpenAlex épinglé.

Notice bibliographique

RevueJournal of electronic commerce research · 2010
Typearticle
Langueen
DomaineComputer Science
ThématiqueSentiment Analysis and Opinion Mining
Établissements canadiensUniversity of Ottawa
Organismes subventionnairesnon disponible
Mots-clésHelpfulnessComputer scienceProduct (mathematics)Ranking (information retrieval)PurchasingThe InternetWorld Wide WebInformation retrievalData scienceMarketingPsychologyBusinessMathematics
DOInon disponible

Résumé

récupéré en direct d'OpenAlex

ABSTRACT Most E-commerce web sites and online communities provide interfaces and platforms for consumers to express their opinions about a specific product by writing personal reviews. The fast development of E-commerce has caused such a huge amount of online product reviews to become available to consumers that it is impossible for potential consumers to read through all the reviews and to make a quick purchasing decision. Review readers are asked to vote if a review is Helpful or Unhelpful and the most positively voted reviews are placed on the top of product review list. However, the accumulation of votes takes time for a review to be fully voted and newly published reviews are always listed at the bottom of the review list. This paper proposes a linear model to predict the helpfulness of online product reviews. Reviews can be quickly ranked and classified by our model and reviews that may help consumers better than others will be retrieved. We compare our model with several machine learning classification algorithms and our experimental results show that our approach effectively classifies online reviews. Also, we provide an evaluation measurement to judge the performance of the helpfulness modeling algorithm and the results show that the helpfulness scores predicted by our approach consistently follow the changing trend of the true helpfulness values. Keywords: recommender system, online product review, helpfulness, evaluation (ProQuest: ... denotes formulae omitted.) 1. Introduction Due to the fast development of Internet and E-commerce, more and more online reviews aggregation web sites, such as Epinions.com etc., have provided consumers with platforms to exchange their opinions about products, services, and merchants. Online product reviews provided by consumers who previously purchased products have become a major information source for consumers and marketers regarding product [Hu & Zhang 2008]. Park et al. [2007] confirmed that the quality of reviews has a positive effect on product sales and consumers purchase intentions increase with the quantity of product reviews. On the E-commerce web sites, such as Amazon.com and Ebay.com, consumers are asked to write reviews and rate products or services by a number of stars after they finished a transaction. Most of existing recommendation approaches [Goldberg et al. 1992; Resnick et al. 1994; Sarwar et al. 2001] are merely based on the rating of products. With a star rating scale, users can not get `real semantics' of review statements. Since product reviews represent reviewers' feelings, experiences and opinions on a specific product, they are more useful than product ratings and therefore can better help potential consumers make purchase decisions. Search engines are good tools to assist in looking for information; however, there are too many search results returned from a search engine and not all of them are reviews. For instance, if we input `Cyber-shot Digital Camera Review' in Google, 278,000 web pages will be returned. This is absolutely a too huge result set for consumers to go through. Moreover, in an online community, such as Epinion.com or Amazon.com, more than 1000 reviews for a specific product are submitted by different consumers. Therefore, it is important to rank and classify product reviews so that they can be accessed easily and used effectively by consumers. Review aggregation web sites provide a function for consumers to vote whether a review is Helpful or Unhelpful. But this progress takes time far before a really helpful review is discovered and the most recent published reviews will always be the least voted ones. Our goal is to develop a model that filters out reviews which are most likely helpful to consumers and that provide more valuable information for consumer's decision making. Such a model can save a great deal of consumers' effort in surfing for reliable and helpful reviews. …

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.

Prédiction distillée sur la base complète

Imitation des enseignants

Ni prévalence calibrée, ni vérité terrain. Validation humaine à venir. Apprise à partir de 10 348 étiquettes directes de Codex et de 10 348 étiquettes directes de Gemma. Le mode candidate est l'union des têtes enseignantes seuillées; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont ni des étiquettes humaines ni des étiquettes directes de modèles de pointe.

score de la tête « metaresearch » (Codex)0,015
score de la tête « metaresearch » (Gemma)0,001
Version: codex-gemma-dda1882f352aStatut de validation: machine_predicted_unvalidated
Catégories candidatesaucune
Catégories consensuellesaucune
DomaineSignal candidat: aucune · Signal consensuel: aucune
Devis d'étudeSignal candidat: Sans objet · Signal consensuel: aucune
GenreSignal candidat: Méthodes · Signal consensuel: aucune
Score de désaccord entre enseignants0,795
Score d'incertitude au seuil0,797

Scores Codex et Gemma par catégorie

CatégorieCodexGemma
Métarecherche0,0150,001
Méta-épidémiologie (sens strict)0,0000,000
Méta-épidémiologie (sens large)0,0010,000
Bibliométrie0,0010,001
Études des sciences et des technologies0,0000,000
Communication savante0,0000,000
Science ouverte0,0020,000
Intégrité de la recherche0,0000,002
Charge utile insuffisante (le modèle a refusé de juger)0,0000,000

Scores machine (provisoires)

Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.

Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.

Tête enseignante Opus0,164
Tête enseignante GPT0,442
Écart entre enseignants0,278 · la distance entre les deux têtes enseignantes sur ce seul travail
Statut de validationscore_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découle