| For potential consumers,online reviews can help them assess the quality of a restaurant and provide important information for their future choices and decisions.However,as the number of online reviews increases and the amount of time and effort required by consumers to process the information contained in the reviews,it becomes important to determine the level of review helpfulness.Factors influencing the helpfulness of reviews,such as ratings,number of reviews,readability of reviews,reviews age and sentiment in reviews,have been studied by scholars.However,it is still worth exploring how the content characteristics of review texts affect their helpfulness using the relatively new information entropy approach.In particular,taking photos and posting reviews with photos becomes increasingly popular,the impact of content information brought by photos on the helpfulness of online reviews can still be empirically analyzed.This study draws on information quality theory and dual processing theory to examine how the information content of review text and review photos affects the helpfulness of reviews on a more granular dimension.While the combination of review text and photo information in online reviews has been shown to have a significant effect on predicting the helpfulness of online reviews,it is still worth investigating which information in the textual content of reviews and photos has a significant effect on predicting the review helpfulness.The study therefore examines the impact of review text and review photos on the review helpfulness based on the characteristics of the reviews,which can be divided into the following two parts:First,a study of the factors influencing the reviews helpfulness based on text analysis.In order to identify the specific information contained in the text content of user reviews to construct the information entropy in the model of influencing the review helpfulness,the study used 59,358 reviews from the Yelp.com website.Based on information quality theory,the concept of information entropy was introduced,and Python software was used to extract word frequencies from the text content in the reviews to obtain restaurant-related food words,service words,and environment words,and then the food entropy,service entropy,and environment entropy of the review texts were calculated based on entropy theory to study the influence of consumer-perceived food,service,and environment information on the review helpfulness.In addition,this paper analyzes the moderating effects of systematic cues and heuristic cues based on the dual processing theory,analyzes the emotional polarity of review texts by using the LIWC sentiment analysis tool(Linguistic Inquiry and Word Count),and introduces visual cues as moderating variables to build a research model for analyzing the influence of review texts on the review helpfulness.The study concluded that food entropy,environmental entropy,and service entropy significantly and positively affect review helpfulness,and visual cues play a positive role in the process of food entropy and service entropy and have the opposite effect in the process of environmental entropy.Positive emotion plays a negative role in the process of environmental entropy and service entropy role in review helpfulness,and negative emotion plays an inhibitory role in the process of food entropy role in review helpfulness.Secondly,a study of the factors influencing the review helpfulness based on image recognition.To investigate the influence of review photo information richness on the review helpfulness,reviews from the Yelp.com website were captured using a web crawler tool,and 5,618 reviews with photos were used after screening.Based on the information quality theory,the study quantifies the visual breadth and visual depth variables by manually identifying the performance of the review photos in each dimension from the food,environment,people,and menu dimensions related to the restaurant,and investigates their impact on the review helpfulness.In addition,the study introduces the NIMA(Neural Image Assessment)model to quantify the aesthetic quality of review photos as a moderating variable and constructs a research model to analyze the influence of review photos on the review helpfulness.The study concludes that visual depth and visual breadth positively contribute to review helpfulness,and photograph aesthetics negatively inhibit the positive effect of visual depth on review helpfulness.The novelty of this study is that it extends the findings of the study on the helpfulness of online restaurant reviews by elucidating the informativeness of review text and photo content based on information quality theory and dual processing theory,thus providing a theoretical basis for the study of the helpfulness of online restaurants.In the study of review content,the content of review text is measured from the relatively new perspective of information entropy,which distinguishes more specifically between food entropy,environmental entropy,and service entropy.The study of review photo distinguishes information richness as the visual depth and visual breadth,revealing the impact of restaurant user-generated photos on the helpfulness of online reviews.This study has important implications for restaurants to better use and manage online reviews.In addition,restaurants can help consumers make better decisions by showing them the most valuable review photos and review text,which is important for improving restaurant quality and increasing business revenue. |
PDF Full Text Request