Search

In the ever-expanding digital landscape, the abundance of user-generated content on consumer platforms such as Booking and TripAdvisor offers a rich source of information for both travellers and hoteliers. Sentiment analysis, a fundamental research task of natural language processing (NLP) is used for mining sentiments and opinions within this vast reservoir of text reviews. A more specific type of sentiment analysis, i.e., aspect-based sentiment analysis (ABSA), is used when processing customer reviews is required. In ABSA, we aim to capture aspect-level sentiments and intricate relationships between various aspects within reviews. This article proposes a novel approach to ABSA by introducing a novel technique of word sense disambiguation (WSD) and integrating it with the Transformer architecture bidirectional encoder representations from Transformers (BERT) and graph convolutional networks (GCNs). The proposed approach resolves the intriguing ambiguities of the words and represents the review data as a complex graph structure, facilitating the modeling of intricate relationships between different aspects. The combination of bidirectional long short-term memory (BiLSTM) and GCN proves effective in capturing inter-dependencies among various aspects, providing a nuanced understanding of customer sentiments. The experiments are conducted on the RABSA dataset (an enhanced and richer hotel review data collection), and results demonstrate that our approach outperforms previous baselines, showcasing the effectiveness of integrating WSD in ABSA. Furthermore, an ablation study confirms the significant contribution of the WSD module to the overall performance. Moreover, we explore different similarity measures and find that cosine similarity yields the best results when identifying the real sense of a word in a given sentence using WordNet. The findings of our work and future work related to our work create lots of interest for people in the tourism and hospitality industry. This research gives another boost to the concept of the potential of NLP techniques in sentiment analysis. It emphasizes that if we combine the potential of NLP techniques along with state-of-the-art machine learning frameworks, we can shape the future of this field. [ABSTRACT FROM AUTHOR]

Indoor Environmental Quality (IEQ) significantly affects occupants' well-being and comfort. Assessing IEQ typically involves post-occupancy evaluation (POE), a method that can be time-consuming and particularly challenging in hotel settings, where guests may be disrupted by frequent requests for feedback. Hence, this paper investigates the capability of text mining to extract valuable information for IEQ assessment, such as identifying the main causes of IEQ dissatisfaction, detecting combined occurrences of IEQ aspects, and exploring the relationship between IEQ dissatisfaction and hotel attractiveness. To this aim, the study analysed 1494 five-star hotels in Europe, comprising 515,738 reviews. Among them, 13.1% contained references to keywords related to IEQ aspects. The major cause of dissatisfaction in hotels is acoustic (42.7% of the reviews), followed by thermal (35.7%), visual (11.1%) comfort, and IAQ (10.5%). Additionally, 9580 reviews demonstrated the co-occurrence of multiple IEQ aspects, highlighting the interplay between different aspects. Furthermore, the reviewer score, reflecting the hotel's attractiveness, showed an inverse relationship with the percentage of dissatisfied guests regarding IEQ, highlighting the impact of the indoor environment on the hotel rating. Overall, text mining is effective in supporting IEQ assessment and the study underscores the effect of addressing IEQ aspects on a facility's overall appeal. [ABSTRACT FROM AUTHOR]

Businesses rely heavily on review analysis of consumer feedback and reviews when making decisions. This research examines fifteen years' worth of hotel evaluations to learn more about the opinions of visitors and how they relate to review scores in various years and regions. After the dataset was cleansed to concentrate on reviews written in English, TextBlob was used to analyze sentiment and classify the reviews as positive, neutral, or negative. The results validate using sentiment analysis in conjunction with numerical ratings, which show a positive link between more excellent ratings and positive sentiment scores. A significant drop in sentiment scores following the 2008-2009 recession emphasizes how the global financial crisis affected the hospitality sector. Significant differences in guest feelings between cities, states, and nations are shown by geographic analysis, underscoring the necessity of region-specific approaches. Sentiment distribution and trends were revealed via intricate visualizations such as heat maps, box plots, scatter plots, and histograms. To improve guest happiness, the study uses an understanding of geographical variations. These results give hotel management a solid framework for raising customer satisfaction and loyalty, customizing marketing campaigns, and enhancing service quality. For deeper insights, future studies may investigate more sophisticated NLP techniques. [ABSTRACT FROM AUTHOR]

Recent research has focused on opinion mining from public sentiments using natural language processing (NLP) and machine learning (ML) techniques. Transformer-based models, such as bidirectional encoder representations from transformers (BERT), excel in extracting semantic information but are resourceintensive. Google's new research, mixing tokens with fourier transform, also known as FNet, replaced BERT's attention mechanism with a non-parameterized fourier transform, aiming to reduce training time without compromising performance. This study fine-tuned the FNet model with a publicly available Kaggle hotel review dataset and investigated the performance of this dataset in both FNet and BERT architectures along with conventional machine learning models such as long short-term memory (LSTM) and support vector machine (SVM). Results revealed that FNet significantly reduces the training time by almost 20% and memory utilization by nearly 60% compared to BERT. The highest test accuracy observed in this experiment by FNet was 80.27% which is nearly 97.85% of BERT's performance with identical parameters. [ABSTRACT FROM AUTHOR]

Despite the prevalence of online reviews, little research has examined the possible linkage between reviewers' communication styles and Hotel Star Class (HSC). Indeed, it is unclear whether review length and communication style change based on the type of hotel being reviewed (proxied by HSC classification). Applying two multivariate analyses of variance to a sample of TripAdvisor® reviews regarding nineteen hotels, this study revealed that less favourable reviews are significantly lengthier than favourable ones, but only for higher-ranked hotels. Furthermore, favourable (vs. less favourable) reviews predominantly reflect affective (vs. cognitive) processes, especially for higher-ranked hotels. The findings underline the importance of considering HSC classification when studying customer contentment in the hospitality industry. [ABSTRACT FROM AUTHOR]

This research tests the relationship between aspects of customer influenceability at the time of booking a hotel with the propensity to write a review in electronic word-of-mouth communication. A valid sample of 739 online questionnaires was obtained. An Exploratory Factor Analysis was conducted in order to reduce the dimensions of the two critical variables, and a measurement model was built. Then a Path analysis was carried out. The novelty of this research lies in measuring the evolution from being a passive eWOM reader to a proactive eWOM writer. Results indicate a relationship between being influenced by reading reviews and the propensity to write reviews. The most important underlying motivation to write a review is altruistic. Managers should try to identify the most responsive customers and encourage them to write reviews on altruistic grounds. This study effectively validated the impact of being responsive to reading reviews on the inclination to, in turn, write them. Findings contribute to the evolving research landscape in eWOM within the hospitality and tourism sector, offering practical insights for industry practitioners to formulate more effective strategies in soliciting and managing customer reviews. [ABSTRACT FROM AUTHOR]

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 andmuch better in terms of performancemetrics. [ABSTRACT FROM AUTHOR]

Introduction. Tourists traveling in the times of crisis have a higher tolerance for risk and can be called 'crisis-resistant'. The fact that they are ready to travel even if it is burdened with additional stress and obstacles makes them a valuable market segment for hoteliers, which is worth striving for particularly in the current pandemic situation. The research aims at better understanding of contemporary 'crisis-resistant' tourists through in-depth analysis of electronic Word of Mouth (eWOM). Material and methods. A web scrapping method was applied to acquire eWOM content posted by tourists traveling in the times of Covid-19. A total of 1239 reviews from 455 hotels were analyzed. Six hypotheses were tested with the use of T Test and Chi Square test. Results. The type of travel and the month of travel did not influence the frequency of coronavirus mentions. Similarly, reviews relating to COVID-19 did not vary in score nor in frequency of managerial replies. However, mention of coronavirus influenced the length of the review and its helpfulness. Conclusions. Crisis-resistant tourists treat issues related to the pandemic (including some in-hotel regulations and restrictions) as any other aspects of the hotel performance and accept them. Interestingly, those aspects do not negatively influence the total evaluation of the hotel, which may suggest that such tourists have already adapted to all the pandemic-induced burdens. At the same time, hotel managers fail to respond timely and adequately to negative comments, which seems to be of the key importance in such a situation. [ABSTRACT FROM AUTHOR]

Copyright of Procesamiento del Lenguaje Natural is the property of Sociedad Espanola para el Procesamiento del Lenguaje Natural and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The Hotel website always provide sentiment reviews of the service, then users can easily choose the desired facility according to the best advice results from the previous user. In this study, the sentiment of the reviews are determined based on five hotel aspects, those are: food, service, location, comfort and cleanliness. Every data hotel reviews is pre-processed to produce a term list. Extract related hidden topics using Latent Dirichlet Allocation (LDA). The extracted documents are categorized by matching for similarity to determine the term documents into five hotel aspects. Each term document is expanded using synonyms to increase the similarity value to LDA with 100% expanded document using Cosine Similarity produces the highest performance value of 0.856. Labeling the sentiment of each review based on the aspect and sentiment classification using the Support Vector Machine method gets an average value of 0.940 for all fifth aspects. The ranking of the most important aspects shows that users talk a lot about the cleanliness aspect, having the highest positive sentiment value of 39.450 and having a negative sentiment of 3.861, indicating that each review sentiment is influenced by certain aspects. [ABSTRACT FROM AUTHOR]

Online customer reviews, being an essential factor that determines success or failure in business, in particular in the tourism industry, demand close attention since the investigation of this type of discourse might bear some implications both for language specialists and for hotel managers. This paper is a quantitative and qualitative study of evaluation devices occurring in the corpus of hotel reviews from Booking.com. The analysis is based on methods associated with corpus linguistics and Systemic Functional Linguistics (Halliday 1994; Eggins 2004), specifically, the Appraisal framework (Martin & White 2005), which is applied here to a new type of discourse, that of hotel ratings. The study focuses on the categories of Attitude, Graduation and Engagement suggested by Martin and White and investigates inscribed and invoked evaluation. In order to adapt the Attitude dimensions to the corpus under study, new subcategories were created as a contribution to evaluation theory. The practical result of the study is a list of evaluation devices employed in hotel reviews; the list may be useful for further corpus analyses and for designing systems for automated analysis of customer evaluation. Since online customer reviews are commonly used by hotels as a benchmark of their guests' satisfaction and a valuable information source of features and services that need to be improved, the present study is also likely to make an important contribution to the industry. [ABSTRACT FROM AUTHOR]

We developed a system for sentiment analysis of hotel reviews. The system uses topics to give accurate reports related to some qualities specified by the end-user. This system permits users to evaluate a given hotel services quality according to multiple criteria. Besides, a user can make different detailed comparisons between hotels. To achieve this goal, the proposed system decomposes complex opinions into sub-comments to extract the specified concept. Further, it translates non-English comments into the English language. After having tested the system by using two datasets, the obtained results are judged to be satisfactory. [ABSTRACT FROM AUTHOR]

Purpose: The purpose of this study is to provide better service to hotel customers during the COVID-19 era. Specifically, this study focuses on understanding the changes in hotel customer satisfaction during the epidemic and formulating effective marketing strategies to satisfy and attract guests. Design/methodology/approach: As the first victim of the COVID-19 virus, China's hotel industry has been profoundly affected and customer satisfaction and needs have also changed. Taking 105,635 hotel reviews obtained from Tripadvisor.com in Beijing and Shanghai as samples, this study explores the changes in consumer satisfaction by using text-mining methods. Findings: The results suggest that there are significant differences in overall ratings, spatial distribution and ratings of different traveller types before and after the epidemic. Generally, customers have higher "tolerance" and are more inclined to give higher ratings and pay more attention to hotel prevention and control measures to reduce health risks after the COVID-19. Research limitations/implications: This paper proves the changes in customer satisfaction before and after the COVID-19 at the theoretical level and reveals the changes in customer attention through the topic model and provides a basis for guiding hotel managers to reduce the impact of the COVID-19 crisis. Practical implications: Empirical findings would provide useful insights into tourism management and improve hotel service quality during the COVID-19 epidemic era. Originality/value: This research explores the hotel customer satisfaction in the field of hotel management before COVID-19 and after COVID-19, by using text mining to analyse mandarin online reviews. The results of this study will suggest that the hotel industry should continuously adjust its products and services based on the effective information obtained from customer reviews, so as to realize the activation and revitalization of the hotel industry in the epidemic era. [ABSTRACT FROM AUTHOR]

Currently, there are a large number of hotel reviews on the Internet that need to be evaluated to turn the data into practicable information. Deep learning has excellent capabilities for recognizing this type of data. With the advances in deep learning paradigms, many algorithms have been developed that can be used in sentiment analysis tasks. In this study, we aim to compare the performance of classical machine learning algorithms-- logistic regression (LR), naïve Bayes (NB), and support vector machine (SVM) using the Word2Vec model in conjunction with deep learning algorithms such as a convolutional neural network (CNN) to classify hotel reviews on the Traveloka website into positive or negative classes. Both learning methods apply hyperparameter tuning to determine the parameters that produce the best model. Furthermore, the Word2Vec model parameters use the skip-gram model, hierarchical softmax evaluation, and the value of 100 vector dimensions. The highest average accuracy obtained was 98.08% by using the CNN with a dropout of 0.2, Tanh as convolution activation, softmax as output activation, and Adam as the optimizer. The findings from the study demonstrate that the integration of the Word2Vec model and the CNN model obtains significantly better accuracy than other classical machine learning methods. [ABSTRACT FROM AUTHOR]

• The theory of construal can be tied to German and English contrasts. • Impersonal pronouns are a form of agent defocusing through delimitation. • German reviewers use impersonal pronouns in diverse generic and episodic contexts. • Impersonal pronouns can impersonalize but also indicate reader relevance. • Contrasts in (im)personalization are a result of typological and linguacultural factors. The current work explores the concept of construal in German and English, focusing specifically on the communicative feature of (im)personalization. Incorporating insights from cognitive linguistics, semantics and pragmatics, it takes as object of analysis a means of im personalization, namely the use of human impersonal pronouns (HIPs) and, using a form-to-function approach, analyzes the dedicated HIPs man and one , as well as impersonally used 2nd personal pronouns du and you regarding their featural composition and pragmatic effects. The data consists of two parallel corpora of negative online hotel reviews taken from Booking.com. The findings reveal that German-speaking reviewers make use of HIPs more frequently and use the 'less personal' of the HIPs (man) as compared to English-speaking reviewers (who use the 'more personal' you), confirming the established contrast of German speakers using more impersonalized language. They further uncover a flexibility in the distribution and usage of German man that is not yet widely established in the literature, showing that German speakers use man in a myriad of ways to simultaneously impersonalize certain aspects of their reviews and indicate relevance to the reader. The findings point to a mixture of typological and linguacultural influences at play regarding the HIPs' role in impersonalization in German and English negative online hotel reviews. [ABSTRACT FROM AUTHOR]

Imbalanced sentiment is one of the key classification issues. Many studies have proposed imbalanced sentiment classification improvements, but the topic remains problematic as a major challenge. This paper proposes a method, called “concept-based one-class SVM classifier”, to address imbalanced sentiment classification that consists of three main techniques. First, we apply Word2Vec and PageRank algorithms to extract “concepts” and their related terms (called “members”) embedded in texts. The corpus of “concepts” is then used to prepare the dataset by replacing words with the “concepts”. This reduces term ambiguity and also the size of word vectors. Second, supervised term weighting (STW) schemes are applied to determine the importance of a word in a document of a specific class. This reflects the class distinguishing power of each term. Finally, the one-class support vector machine (SVM) algorithm is used for sentiment classifier modeling. This has proved useful for imbalanced data classification, especially when the minority class lacks structure and is predominantly composed of small disjuncts or outliers. By combining these techniques, our proposed method may be able to competently identify and distinguish between the characteristics of each class, especially in the context of an imbalanced data scenario. After validating the proposed method with the hotel review dataset, and running experiments with different imbalanced ratios, our proposed method returned satisfactory results of recall, precision, and F1. We then selected the best model generated from our method and compared the results to the state-of-the-art method. Our proposed method returned better results than the state-of-the-art method, with improved scores of F1 at 3.19%. Moreover, if considering for the computational processing time, our proposed method is faster than the state-of-the-art method. [ABSTRACT FROM AUTHOR]

Purpose: This study aims to explore US hotel guests' privacy concerns with a twofold aim as follows: to investigate the privacy categories, themes and attributes most commonly discussed by guests in their reviews and to examine the influence of cultural proximity on privacy concerns. Design/methodology/approach: This study combined automated text analytics with content analysis. The database consisted of 68,000 hotel reviews written by US guests lodged in different types of hotels in five European cities. Linguistic Inquiry Word Count, Leximancer and SPSS software were used for data analysis. Automated text analytics and a validated privacy dictionary were used to investigate the reviews by exploring the categories, themes and attributes of privacy concerns. Content analysis was used to analyze the narratives and select representative snippets. Findings: The findings revealed various categories, themes and concepts related to privacy concerns. The two most commonly discussed categories were privacy restriction and outcome state. The main themes discussed in association with privacy were "room," "hotel," "breakfast" and several concepts within each of these themes were identified. Furthermore, US guests showed the lowest levels of privacy concerns when staying at American hotel chains as opposed to non-American chains or independent hotels, highlighting the role of cultural proximity in privacy concerns. Practical implications: Hotel managers can benefit from the results by improving their understanding of hotel and service attributes mostly associated with privacy concerns. Specific suggestions are provided to hoteliers on how to increase guests' privacy and on how to manage issues related to cultural distance with guests. Originality/value: This study contributes to the hospitality literature by investigating a neglected issue: on-site hotel guests' privacy concerns. Using an unobtrusive method of data collection and text analytics, this study offers valuable insights into the categories of privacy, the most recurrent themes in hotel guests' reviews and the potential relationship between cultural proximity and privacy concerns. [ABSTRACT FROM AUTHOR]

Generally, Online Travel Agent (OTA) has a review element where clients can give reviews of the facilities they have used. Availability of a huge volume of reviews makes it troublesome for service executives to know the percent of reviews that have an effect on their services. Thus, it is essential to develop a sentiment assessment technique with respect to hotel reviews, particularly in Indonesian language. This research makes use of Long-Short Term Memory (LSTM) model as well as the Word2Vec model. The integration of Word2Vec and LSTM variables used in this research are Word2Vec architecture, Word2Vec vector dimension, Word2Vec evaluation method, pooling technique, dropout value, and learning rate. On the basis of an experimental research performed through 2500 review texts as dataset, the best performance was obtained that had accuracy of 85.96%. The parameter combinations for Word2Vec are Skip-gram as architecture, Hierarchical Softmax as evaluation method, and 300 as vector dimension. Whereas the parameter combinations for LSTM are dropout value is 0.2, pooling type is average pooling, and learning rate is 0.001. [ABSTRACT FROM AUTHOR]

The eWOM has been widely used by scholars and the hospitality industry as a proxy for customers' opinions. However, only a handful of studies examined customers' opinions on different review platforms, which questions the transferability of results obtained from one platform to another. In a systematic investigation, we found that reviews of the same hotels from major eWOM platforms differ in leading eWOM indicators: star ratings, topics, and sentiment, especially between the closed and open review platforms. Over time, a "U-shape" review score distribution emerges signaling bias toward the opinion extremes. The study provides strong evidence for academics and hoteliers not to rely on a single platform to evaluate customer experiences, satisfaction, marketing effectiveness, and the overall image of their brand. We discuss the potential reasons for platform divergence and research avenues to improve our knowledge of inter-platform review variability and develop more rigorous methods employing social media data. • We analyzed the scores, sentiments, and topics of 75,000 hotel reviews from four platforms. • Topics, sentiments, ratings, and other eWOM metrics of the same hotels differ among review platforms. • We observed the highest eWOM contrast between the open and closed review sites. • Over time, a U shape distribution of review metrics emerges showing bias toward opinion extremes. • Academics and hoteliers should use multiple platforms to evaluate customer experiences. [ABSTRACT FROM AUTHOR]

Identification of hotel competitiveness is crucial for hotel managers in developing effective marketing strategies and promoting their business. A common practice in representing hotel competitiveness is to advertise hotel features with high customer ratings as selling points. However, this approach is ineffective to clearly distinguish top-tier hotels with similar high ratings, while low-tier hotels are disadvantaged because of their low ratings. This study presents a new approach for hotel competitiveness evaluation by identifying the unique aspects that combine multiple features. The proposed approach could distinguish a hotel from its counterparts, rather than rely solely on the high ratings of hotel features. We introduce a data mining technique to automatically discover the distinct aspects of a hotel based on the ratings of hotel features. The proposed method is beneficial to researchers and hotel managers in evaluating hotel competitiveness and identifying critical features that need improvement to considerably attract potential customers. [ABSTRACT FROM AUTHOR]

Finding a suitable hotel based on user's need and affordability is a complex decision-making process. Nowadays, the availability of an ample amount of online reviews made by the customers helps us in this regard. This very fact gives us a promising research direction in the field of tourism called hotel recommendation system which also helps in improving the information processing of consumers. Real-world reviews may showcase different sentiments of the customers towards a hotel and each review can be categorized based on different aspects such as cleanliness, value, service, etc. Keeping these facts in mind, in the present work, we have proposed a hotel recommendation system using Sentiment Analysis of the hotel reviews, and aspect-based review categorization which works on the queries given by a user. Furthermore, we have provided a new rich and diverse dataset of online hotel reviews crawled from Tripadvisor.com. We have followed a systematic approach which first uses an ensemble of a binary classification called Bidirectional Encoder Representations from Transformers (BERT) model with three phases for positive–negative, neutral–negative, neutral–positive sentiments merged using a weight assigning protocol. We have then fed these pre-trained word embeddings generated by the BERT models along with other different textual features such as word vectors generated by Word2vec, TF–IDF of frequent words, subjectivity score, etc. to a Random Forest classifier. After that, we have also grouped the reviews into different categories using an approach that involves fuzzy logic and cosine similarity. Finally, we have created a recommender system by the aforementioned frameworks. Our model has achieved a Macro F1-score of 84% and test accuracy of 92.36% in the classification of sentiment polarities. Also, the results of the categorized reviews have formed compact clusters. The results are quite promising and much better compared to state-of-the-art models. The relevant codes and notebooks can be found here. • Designed a Hotel Recommendation System based on online review in English language. • Ensemble of BERT and Random Forest models for classifying sentiments on reviews. • Used textual features like Word2Vec embeddings and TF–IDF scores. • Categorized the reviews based on aspects using Fuzzy logic and Cosine similarity. • Prepared a sentiment tagged dataset from Tripadvisor consisting of hotel reviews. [ABSTRACT FROM AUTHOR]