Hotel Reviews Research Articles

The value of rating diversity within multidimensional rating system: Evidence from hotel booking platform

An increasing number of hotel booking platforms and review websites deploy multidimensional rating systems to encourage users to provide additional evaluations of a product/service besides the overall rating. However, providing the itemized ratings of a product/service requires extra effort which makes users tend to post identical ratings (default setting). According to the information transfer theory, we postulate that rating diversity enhances review usefulness within the multidimensional rating system and; based on the loss aversion phenomenon, we also posit that the positive effect of rating diversity on review usefulness should be greater among negative reviews. Using a sample of 1,720,429 hotel reviews, this study reveals that while rating diversity enhances review usefulness, its effect varies across the valence of each review; specifically, the positive effect of rating diversity mainly functions among negative reviews. These findings yield direct implications for hotel booking platforms or review websites that deploy multidimensional rating systems.

Semantic Features-Based Discourse Analysis Using Deceptive and Real Text Reviews

Social media usage for news, feedback on services, and even shopping is increasing. Hotel services, food cleanliness and staff behavior are also discussed online. Hotels are reviewed by the public via comments on their websites and social media accounts. This assists potential customers before they book the services of a hotel, but it also creates an opportunity for abuse. Scammers leave deceptive reviews regarding services they never received, or inject fake promotions or fake feedback to lower the ranking of competitors. These malicious attacks will only increase in the future and will become a serious problem not only for merchants but also for hotel customers. To rectify the problem, many artificial intelligence–based studies have performed discourse analysis on reviews to validate their genuineness. However, it is still a challenge to find a precise, robust, and deployable automated solution to perform discourse analysis. A credibility check via discourse analysis would help create a safer social media environment. The proposed study is conducted to perform discourse analysis on fake and real reviews automatically. It uses a dataset of real hotel reviews, containing both positive and negative reviews. Under investigation is the hypothesis that strong, fact-based, realistic words are used in truthful reviews, whereas deceptive reviews lack coherent, structural context. Therefore, frequency weight–based and semantically aware features were used in the proposed study, and a comparative analysis was performed. The semantically aware features have shown strength against the current study hypothesis. Further, holdout and k-fold methods were applied for validation of the proposed methods. The final results indicate that semantically aware features inspire more confidence to detect deception in text.

Weak tagging and imbalanced networks for online review sentiment classification

Sentiment classification aims to complete the automatic judgment task of text sentiment tendency. In the sentiment classification task of online reviews, traditional deep learning models require a large number of manually annotated samples of sentiment tendency for supervised training. Faced with massive online review data, the feasibility of manual tagging is worrisome. In addition, the traditional deep learning model ignores the imbalanced distribution of the number of classification samples, which will lead to a decline in classification performance in the practical application of the model. Considering that the online review data contains weak tagging information such as scores and labels, and the distribution is imbalanced, a weak tagging and imbalanced networks for online review sentiment classification is constructed. The experimental results show that the model significantly outperforms the traditional deep learning model in the sentiment classification task of hotel review data.

Information extraction and sentiment analysis of hotel reviews in Croatia

Today, the amount of data in and around the business system requires new ways of data collection and processing. Discovering sentiments from hotel reviews helps improve hotel services and overall online reputation, as potential guests largely consult existing hotel reviews before booking. Therefore, hotel reviews of Croatian hotels (categories three, four, and five stars) in tourist regions of Croatia were studied on the Booking.com platform for the years 2019 and 2021 (before and after the start of the pandemic COVID-19). Hotels on the Adriatic coast were selected in the cities that were mentioned by several sources as the most popular: Rovinj, Pula, Krk, Zadar, Šibenik, Split, Brač, Hvar, Makarska, and Dubrovnik. The reviews were divided into four groups according to the overall rating and further divided into positive and negative in each group. Therefore, the elements that were present in the positive and negative reviews of each of the four groups were identified. Using the text processing method, the most frequent words and expressions (unigrams and bigrams), separately for the 2019 and 2021 tourism seasons, that can be useful for hotel management in managing accommodation services and achieving competitive advantages were identified. In the second part of the work, a machine learning (ML) model was built over all the collected reviews, classifying the reviews into positive or negative. The results of applying three different ML algorithms with precision and recall performance are described in the Results and Discussion section.

An Ensemble-Based Hotel Reviews System Using Naive Bayes Classifier

The task of classifying opinions conveyed in any form of text online is referred to as sentiment analysis. The emergence of social media usage and its spread has given room for sentiment analysis in our daily lives. Social media applications and websites have become the foremost spring of data recycled for reviews for sentimentality in various fields. Various subject matter can be encountered on social media platforms, such as movie product reviews, consumer opinions, and testimonies, among others, which can be used for sentiment analysis. The rapid uncovering of these web contents contains divergence of many benefits like profit-making, which is one of the most vital of them all. According to a recent study, 81% of consumers conduct online research prior to making a purchase. But the reviews available online are too huge and numerous for human brains to process and analyze. Hence, machine learning classifiers are one of the prominent tools used to classify sentiment in order to get valuable information for use in companies like hotels, game companies, and so on. Understanding the sentiments of people towards different commodities helps to improve the services for contextual promotions, referral systems, and market research. Therefore, this study proposes a sentiment-based framework detection to enable the rapid uncovering of opinionated contents of hotel reviews. A Naive Bayes classifier was used to process and analyze the dataset for the detection of the polarity of the words. The dataset from Datafiniti’s Business Database obtained from Kaggle was used for the experiments in this study. The performance evaluation of the model shows a test accuracy of 96.08%, an F1-score of 96.00%, a precision of 96.00%, and a recall of 96.00%. The results were compared with state-of-the-art classifiers and showed a promising performance and much better in terms of performance metrics.

Domain Adaptation Approach for Arabic Sarcasm Detection in Hotel Reviews based on Hybrid Learning

Posting sarcastic comments on hotel opinion websites became a common trend. Therefore, sarcasm detection becomes of primary importance, mainly since sarcasm can flip the review's polarity. To our knowledge, no work related to detecting sarcasm exists in the hospitality industry. The lack of a labeled dataset is an utmost issue for researchers. We propose a new approach based on the domain adaptation using the self-training technique for the semi-supervised learning of sarcastic Arabic hotel reviews. We use AraBERT for contextual embedding and machine learning models for self-training classification. The obtained accuracy reaches 91.1%.

A Conceptual and Computational Framework for Aspect-Based Collaborative Filtering Recommender Systems

Many datasets in E-commerce have rich information about items and users who purchase or rate them. This information can enable advanced machine learning algorithms to extract and assign user sentiments to various aspects of the items thus leading to more sophisticated and justifiable recommendations. However, most Collaborative Filtering (CF) techniques rely mainly on the overall preferences of users toward items only. And there is lack of conceptual and computational framework that enables an understandable aspect-based AI approach to recommending items to users. In this paper, we propose concepts and computational tools that can sharpen the logic of recommendations and that rely on users’ sentiments along various aspects of items. These concepts include: The sentiment of a user towards a specific aspect of a specific item, the emphasis that a given user places on a specific aspect in general, the popularity and controversy of an aspect among groups of users, clusters of users emphasizing a given aspect, clusters of items that are popular among a group of users and so forth. The framework introduced in this study is developed in terms of user emphasis, aspect popularity, aspect controversy, and users and items similarity. Towards this end, we introduce the Aspect-Based Collaborative Filtering Toolbox (ABCFT), where the tools are all developed based on the three-index sentiment tensor with the indices being the user, item, and aspect. The toolbox computes solutions to the questions alluded to above. We illustrate the methodology using a hotel review dataset having around 6000 users, 400 hotels and 6 aspects.

The main aspects of the experience economy in tourism industry

The purpose of the study is to identify the influence of experience economy factors (entertainment, education, aesthetics, escapism) and the economic component on preferences in choosing hotels in the domestic tourism industry. The empirical base of the study includes 30,926 reviews from the tripadvisor.ru website for 1,243 hotels in Russia for 2021–2022, divided by consumer ratings. The research methodology includes textual analysis of tourist reviews: counting particular words and phrases in reviews, calculating the correlation of words, finding word relationships using the word2vec embedding matrix, highlighting the most popular topics of discussion based on the Latent Dirichlet allocation (LDA) method and assessing sentiment using a BERT transformer network. The results of this study assess aspects of the basic model of the experience economy (aesthetics, escapism, entertainment and education) and adapt the model for the domestic tourism and hospitality industry, adding relevant aspects (recommendations and economics). Based on the results of a quantitative assessment of aspects of the extended model of the experience economy, it was revealed that the aesthetic aspect is decisive for people when evaluating hotels, and the aspects of entertainment and education, expressed indirectly, although they are of high importance, but not as much as aesthetic parameters. The economic component, primarily the price-quality ratio and the overall assessment of the cost, also stands out in the main topics, however, hotel reviews focus on it to a lesser extent. This is the first time in Russia that such a study is evaluating aspects of the experience economy model for the tourism industry and will allow developing the travel market in Russia.

The Comparison of Language Models with a Novel Text Filtering Approach for Turkish Sentiment Analysis

Today, comments can be made on many topics on web platforms with the development of the internet. Analyzing the data of these comments is essential for companies and data scientists. There are many methods for analyzing data. Recently, language models have also been used in many studies for sentiment analysis or text classification. In this study, Turkish sentiment analysis is performed using language models on hotel and movie review datasets. The language models are chosen because they are rarely used in Turkish literature. The pre-trained BERT, ALBERT, ELECTRA, and DistilBERT models for the Turkish language are trained and tested with these datasets. In addition, a text filtering method, which removes the words that can provide the opposition sentiment in the positive or negative labeled text, is proposed for sentiment analysis. These datasets obtained by this method are also retrained with language models and the accuracy values of their models are measured. The results of this study are compared with previous studies using the same datasets. As a result of the analysis, the accuracy values obtain state-of-the-art results with language models compared to previous studies. The best performance has been achieved by training the ELECTRA language model using the proposed text filtering method.

A novel hybrid deep learning model for aspect based sentiment analysis

SummaryThe usage of social media, forums, and e‐commerce websites have been widely increased. Feedback from customers has a big impact on the final product. A service provider, merchant, or manufacturer need all the information, even if it is just a comment or a review about a service or a product. So, it is vital to look at input from users, and therefore sentiment analysis has received a lot of interest. Sentiment analysis is a method for identifying and analyzing text in order to determine the features, qualities, and viewpoints of particular user. Extracting user aspects is the main part of this process, and it is used to group the user aspects. In recent years, convolutional neural network (CNN) models have gained popularity in natural language processing. Thus, this research proposes a novel hybrid CNN model by concatenating the bidirectional long short‐term memory and CNN models to process the data sequentially by learning their high‐level features. The concatenated method minimizes the loss of critical information. Benchmark product reviews and hotel review datasets are employed in the experiments, and accuracies of 93.6% for the product review dataset and 92.7% for the hotel review dataset are achieved by the proposed hybrid model when compared to state‐of‐the‐art techniques.

Peran Brand Image Situs Tripadvisor Memediasi Pengaruh Ewom terhadap Niat Beli Kamar di Kabupaten Badung

EWOM given by other consumers in a sharing review platform or community can influence consumers' purchase intentions. TripAdvisor as an online review site that provides access for its customers to write a review. On the TripAdvisor site, people who want to share their experiences are required to become regular users and have their TripAdvisor account first and it is known that 65% of respondents see hotel reviews before making a booking. Thus, high ratings and good reviews from customers can create profits for hoteliers as EWOM can help hotels attract new customers. The purpose of this study was to determine the effect of EWOM on room purchase intention in Badung Regency and the role of Tripadvisor website brand image in mediating the relationship between EWOM and room purchase intention. This research method uses a quantitative approach and Partial Least Square Structural Equation Modeling (PLS-SEM) analysis. The population determined in this study is the Indonesian population with a sample of 180 respondents in accordance with the provisions of the sample calculation. The equation models analyzed are the outer model, the inner model and hypothesis testing. Based on the results of calculations using PLS-SEM software, the results show that there is no significant effect between the effect of EWOM on room purchase intentions, there is a significant effect of brand image on EWOM, the brand image of the Tripadvisor site has an effect on room purchase intention and there is a significant influence on brand image the Tripadvisor website as a media to mediate the influence of EWOM on room purchase intentions in Badung Regency. The implication of the results of this study shows that hotels around Badung Regency must maintain professionalism in improving the quality of services and facilities provided to tourists so that they can increase the value of reviews on the Tripadvisor site.

Arabic aspect sentiment polarity classification using BERT

Aspect-based sentiment analysis (ABSA) is a textual analysis methodology that defines the polarity of opinions on certain aspects related to specific targets. The majority of research on ABSA is in English, with a small amount of work available in Arabic. Most previous Arabic research has relied on deep learning models that depend primarily on context-independent word embeddings (e.g. word2vec), where each word has a fixed representation independent of its context. This article explores the modeling capabilities of contextual embeddings from pre-trained language models, such as BERT, and making use of sentence pair input on Arabic aspect sentiment polarity classification task. In particular, we develop a simple but effective BERT-based neural baseline to handle this task. Our BERT architecture with a simple linear classification layer surpassed the state-of-the-art works, according to the experimental results on three different Arabic datasets. Achieving an accuracy of 89.51% on the Arabic hotel reviews dataset, 73.23% on the Human annotated book reviews dataset, and 85.73% on the Arabic news dataset.

Recognize The Polarity of Hotel Reviews using Support Vector Machine

A brand is very dependent on consumer perceptions of the product or services. In assessing consumer perceptions of products and services, companies are often faced with data analysis problems. One of the data that is very useful to produce a picture of consumer perceptions of the products and services is review data. So that the company's ability to process review data means that the company has a picture of the strength of the brand it has. Some of the most popular machine learning algorithms for creating text classification models include the naive Bayes family of algorithms, support vector machines (SVM) and deep learning algorithms. In this research, SVM has been proven to be a reliable method in pattern recognition. In particular, this study aims to produce a model that can be used to classify the polarity of hotel reviews automatically. The experimental data comes from review data on hotels in Europe sourced from TripAdvisor with a total of 38000 reviews. We also measure the quality of the classification engine model. The test results of the SVM model built from hotel review data are quite good. The average accuracy of the classification engine is 92.48%. Because the recall and precision values ​​are balanced, the accuracy value is considered sufficient to describe the quality of the classification.

Aspect-based sentiment analysis for hotel reviews using an improved model of long short-term memory

Advances in information technology have given rise to online hotel reservation options. The user review feature is an important factor during the online booking of hotels. Generally, most online hotel booking service providers provide review and rating features for assessing hotels. However, not all service providers provide rating features or recap reviews for every aspect of the hotel services offered. Therefore, we propose a method to summarise reviews based on multiple aspects, including food, room, service, and location. This method uses long short-term memory (LSTM), together with hidden layers and automation of the optimal number of hidden neurons. The F1-measure value of 75.28% for the best model was based on the fact that (i) the size of the first hidden layer is 1,200 neurons with the tanh activation function, and (ii) the size of the second hidden layer is 600 neurons with the ReLU activation function. The proposed model outperforms the baseline model (also known as standard LSTM) by 10.16%. It is anticipated that the model developed through this study can be accessed by users of online hotel booking services to acquire a review recap on more specific aspects of services offered by hotels

Explaining customer satisfaction via hotel reviews: A comparison between pre- and post-COVID-19 reviews

This study aims to examine whether and how COVID-19 has changed the effects of consumer evaluations of hotel attributes on customer satisfaction. We extracted positive and negative evaluations of hotel attributes from online reviews both pre- and post-COVID-19 and examined their effects on customer satisfaction. Using a sample of 1,947,391 reviews of 35,022 Chinese hotels collected from ctrip.com, we conducted a fine-grained sentiment analysis based on sentiment triples to identify important positive and negative evaluations of hotel attributes. Subsequently, we applied regression analyses to examine how these evaluations of hotel attributes influenced customer satisfaction. The results revealed that positive and negative evaluations of hotel attributes had differentiated effects on customer satisfaction. We classified these attributes into basic, excitement, and performance attributes, from which management implications can be derived.

How do online review valence and ratings interact with consumer-generated visuals?

ABSTRACT The growth of social media and technology has given online reviews more importance and popularity. Consumer-generated visuals (pictures and videos), together with words and numerical components, are increasingly being used in online reviews. However, more research is necessary to understand how these components interact. This study aims to examine the relationships between review valence, numerical ratings, and hotel booking intentions, and investigate the interactions between consumer-generated visuals and demographics on these relationships. An online questionnaire was used to collect data using a convenience sample of 418 customers from Oman. The proposed model was tested using Structural Equation Modeling. The results demonstrated that negative review valence, positive review valence, and rating usefulness are all significant predictors of hotel booking intentions. The results also show that young and female customers are more affected by review valence and rating usefulness. Consumer-generated visuals play a moderating role, where the relationships between hotel booking intentions and review valence and ratings are weaker when customers are attentive to visuals. The study’s results underline the role of negative valence, rating usefulness and visuals, and offer theoretical and practical implications.

Multi-Label Annotation and Classification of Arabic Texts Based on Extracted Seed Keyphrases and Bi-Gram Alphabet Feed Forward Neural Networks Model

In natural language processing, text classification is a fundamental problem. Multi-label classification of textual data is a challenging topic in text classification where an instance can be associated with more than one label. This paper presents a multi-label annotation and classification methodology for Arabic text data that is not currently classified as multi-label, aiming to analyze and compare the performance of various multi-label learning approaches. The current work includes two phases: The first involves automatic annotation of hotel reviews with more than one label based on the aspects found in the reviews. In this phase, review data instances were automatically annotated as multi-label based on the extracted seed keyphrases clusters. The second phase involves experiments to compare the performance of various multi-label classification learning methods. In this phase, we introduced different models including a feed-forward networks model that learns a vector representation based on the bi-gram alphabet rather than the commonly used bag-of-words model. The bi-gram alphabet vector representation model has the advantage of having reduced feature dimensions and not requiring natural language processing tools. The results indicated that employing the bi-gram alphabet vector representation feed forward neural network is a competitive solution for the multi-label text classification problem. It has achieved an accuracy of about 75.2%, and standard deviation (0.062).

Chinese Text Sentiment Classification Based on ERNIE and BiLSTM-AT

Chinese text sentiment classification is a sub-task of natural language processing. However, when text representation is carried out, the polysemy of a word cannot be processed when using the traditional language model to construct the word vector, and the long-distance text information cannot be fully extracted when extracting text features. To solve this problem, this paper proposes a text sentiment classification model combining ERNIE and BiLSTM-AT. First, the pre-training model ERNIE is used to obtain the word vector representation of the fused statement context. Then, the bidirectional long-short-term memory neural network is used to extract the context information and depth semantic information of the text. Then, the attention mechanism is used to assign the corresponding weights to the hidden layer vectors of each time step output by the BiLSTM layer, and the weighted summation is integrated into the sentence features. Finally, the softmax function is used to calculate the probability distribution of the emotional category of the text in the output layer. The results show that the proposed model can achieve high accuracy on both hotel reviews and takeaway reviews. Based on the pre-training model, adding bidirectional long-term and short-term memory network and attention mechanism is beneficial to improve the classification effect of the model, and has certain practicability in text sentiment classification tasks.

The Determinants of Helpful Hotel Reviews: A Social Influence Perspective

Online hotel reviews regarding specific experiences and the sensibility of hotel reviewers are important sources of information for consumers who want to book hotels in the future. However, since thousands of reviews are written for one hotel, it is practically impossible for customers to read all the reviews. To alleviate this problem, OTAs (online travel agencies) provide a helpful vote of reviews, helping customers to quickly find helpful reviews without much effort. Therefore, in this study, a ZINB (zero-inflated negative binomial regression) was performed to investigate factors that influence the helpfulness of hotel reviews using the social influence theory. As a result of the analysis, it was found that location, service, and value of the hotel, alongside normative influencing factors and the length of the review (including informational influencing factors), affect the helpfulness of the review regardless of the city. The results of this study are expected to help hotel managers to take remedial action on negative reviews and strengthen the promotion of positive reviews, potentially helping to increase customer satisfaction.

Role of Trusting Beliefs and Trust in the Adoption of Online Reviews of Hotels : Extension of the IAM Model

This study examined the role of trusting beliefs (TB) and trust in online reviews of hotels. It proposed to extend the information adoption model (IAM) to describe the usefulness of online reviews. The other factors examined were source credibility (SC), relevance, and accuracy. The study’s conceptual model was tested on 208 respondents using structural equation modeling (SEM). The study investigated the precursors to information usefulness (IU), which was strongly associated with information adoption (IA). It revealed that trust in online reviews had a strong association with their IU, but there was no direct relationship between the SC of these reviews and IU. SC was associated with trust, which was associated with IU. TB was also strongly associated with SC. Relevance was associated with IU, but accuracy was not associated with the same. .