Ziqiong Zhang

Sentiment classification of online reviews to travel destinations by supervised machine learning approaches

The rapid growth in Internet applications in tourism has lead to an enormous amount of personal reviews for travel-related information on the Web. These reviews can appear in different forms like BBS, blogs, Wiki or forum websites. More importantly, the information in these reviews is valuable to both travelers and practitioners for various understanding and planning processes. An intrinsic problem of the overwhelming information on the Internet, however, is information overloading as users are simply unable to read all the available information. Query functions in search engines like Yahoo and Google can help users find some of the reviews that they needed about specific destinations. The returned pages from these search engines are still beyond the visual capacity of humans. In this research, sentiment classification techniques were incorporated into the domain of mining reviews from travel blogs. Specifically, we compared three supervised machine learning algorithms of Naive Bayes, SVM and the character based N-gram model for sentiment classification of the reviews on travel blogs for seven popular travel destinations in the US and Europe. Empirical findings indicated that the SVM and N-gram approaches outperformed the Naive Bayes approach, and that when training datasets had a large number of reviews, all three approaches reached accuracies of at least 80%.

Sentiment classification of Internet restaurant reviews written in Cantonese

Research highlights? Naive Bayes and SVM are used for Cantonese sentiment classification. ? Accuracy is influenced by interaction between classification models and features. ? Naive Bayes classifier achieves as well as or better accuracy than SVM. ? Character-based bigrams are better features than unigrams and trigrams in capturing Cantonese sentiment. Cantonese is an important dialect in some regions of Southern China. Local online users often represent their opinions and experiences on the web with written Cantonese. Although the information in those reviews is valuable to potential consumers and sellers, the huge amount of web reviews make it difficult to give an unbiased evaluation to a product and the Cantonese reviews are unintelligible for Mandarin Chinese speakers.In this paper, standard machine learning techniques naive Bayes and SVM are incorporated into the domain of online Cantonese-written restaurant reviews to automatically classify user reviews as positive or negative. The effects of feature presentations and feature sizes on classification performance are discussed. We find that accuracy is influenced by interaction between the classification models and the feature options. The naive Bayes classifier achieves as well as or better accuracy than SVM. Character-based bigrams are proved better features than unigrams and trigrams in capturing Cantonese sentiment orientation.

Effects of managerial response on consumer eWOM and hotel performance: Evidence from TripAdvisor

Purpose This study aims to measures the effects of managerial response on consumer electronic word-of-mouth (eWOM) and hotel performance. Design/methodology/approach A sample of 56,284 consumer reviews and 10,793 managerial responses for 1,045 hotels was retrieved from TripAdvisor, along with 30,232 performance records matched to these hotels on a quarterly basis. Findings This study finds that managerial response leads to an average increase of 0.235 stars in the TripAdvisor ratings of the sampled hotels, as well as a 17.3 per cent increase in the volume of subsequent consumer eWOM. Moreover, managerial response moderates the influence of ratings and volume of consumer eWOM on hotel performance. Practical implications This study offers a practical model that enables hotel managers to orchestrate social media marketing approaches and efforts toward an optimal social media strategy. Originality/value This study differs from extant literature that has extensively focused on consumer reviews by providing a new perspective of management intervention in the social media context. By examining the interplay of managerial response and consumer eWOM at the individual hotel level, this study provides empirical evidence of managerial response affecting hotel performance through the increased ratings and volume of consumer eWOM. This study also offers insights into the practical importance of crafting intervention opportunities to cultivate the continued engagement of consumers on social media and increased hotel performance.

Sentiment classification of online Cantonese reviews by supervised machine learning approaches

Cantonese is an important Chinese dialect spoken in some regions of Southern China. Local online users often represent their opinions and experiences with written Cantonese on the web. With two supervised machine learning approaches, this paper conducts a series of experiments to explore appropriate methods for automatic sentiment classification in the very noisy domain of online Cantonese-written reviews. Findings indicate that the support vector machine classifier based on a Mandarin Chinese word segmentation tool performs surprisingly well. The accuracy, precision and recall respectively for positive and negative reviews all reach above 85% when the training corpus contains 5,000 or more reviews.

SNIPPET-BASED UNSUPERVISED APPROACH FOR SENTIMENT CLASSIFICATION OF CHINESE ONLINE REVIEWS

Sentiment classification seeks to identify general attitude of a piece of text of comments or reviews on certain subject, be it positive or negative. Most existing researches on sentiment classification employ supervised learning approaches that rely on annotated data. However, sentiment is expressed differently on different subjects in different domains, and having annotated corpora for every domain of interest is not always practical. This paper proposes an unsupervised learning approach for classifying text of online reviews as recommended or not recommended. The proposed method is based on search engine snippet, summary information on the result page of a search engine. A basic assumption is that terms with similar orientation tend to co-occur. The co-occurrence is measured by utilizing snippets returned from search engines, with a query consisting of the text and a seed positive or negative word. With the information of snippets, the proposed method may estimate the association of candidate terms more accurately. This allows us to reliably predict the sentiment orientation of customer reviews. Texts of customer reviews are then classified as recommended or not recommended if the average sentiment orientations of its phrases are positive or negative. The research data set of this study consists of 600 Chinese online reviews about travel destinations retrieved from Ctrip.com. Our approach achieves an accuracy of 76.5%. Factors that influence the accuracy of the sentiment classification of Chinese online reviews were discussed.

Differences and similarities in perceptions of hotel experience : the role of national cultures

ABSTRACT This study focuses on demonstrating the differences and similarities between and among Western and Asian customers in terms of travelers’ evaluation of satisfaction, perceived value, and attribute quality in a Hong Kong hotel setting. It has been noted that national culture has a substantial impact on travelers’ expectations and perceptions. The results show that satisfaction, perceived value, and hotel attribute evaluations are significantly higher for Western than Asian travelers. We also found considerable similarities between Australian, Canadian, and British travelers and also between Singaporean and Chinese travelers. In addition, American travelers’ ratings are closest to the average of the total sample compared with travelers from other countries, while Japanese travelers demonstrate the highest expectations and needs among the target nations.

Automatic Detection of Subjective Sentences Based on Chinese Subjective Patterns

Subjectivity analysis requires lists of subjective terms and corpus resources. Little work to date has attempted to automatically recognize subjective sentences and create corpus resources for Chinese subjectivity analysis. In this paper, we present a bootstrapping process that can use subjective phrases to automatically create training set from unannotated data, which is then fed to a subjective phrases extraction algorithm. The learned phrases are then used to identify more subjective sentences. The bootstrapping process can learn many subjective sentences and phrases. We show that the recall for subjective sentences is increased with slightly drop in reliability.

Consumer motivation in providing high-quality information: building toward a novel design for travel guide websites

ABSTRACT Given the ever-growing information overload among users of online review websites, understanding the manner in which cognitive costs are reduced has become increasingly important. Despite the emergence of literature that focused on studying consumer motivation to publish high-quality reviews, the effect of direct intervention on this motivation remains relatively unclear. This study targets a unique design based on data from Qunar.com, one of the most popular Chinese travel websites. By developing a conceptual framework and applying survival analysis methods, our results suggest that both the immediate and cumulative effects of expert reviews positively motivate users to post additional eligible reviews; however, their motivation to further improve the quality of their evaluations diminish as they accumulate more expert reviews. Based on the study results, important implications for travel-related websites are presented.

The Impact of Submission Experiences on Perceptions of Journal Quality and Editorial Support: The Viewpoint of Chinese Authors

This study examines differences in perceptions of journal quality and editorial support among three categories of Chinese authors those whose manuscripts were accepted without revision, those whose manuscripts were accepted after revision, and those whose manuscripts were rejected. An analysis of online reviews of journal quality and editorial support in six disciplines revealed the existence of biases caused by authors’ submission experiences. The results show that a Chinese author will rate the quality of a journal and its editorial support higher if his or her manuscript was accepted by the journal regardless of whether he or she was required to make revisions. The results also indicate that no major variations exist in perceptions of journal quality and editorial support between authors whose manuscripts were accepted without revision and authors whose manuscripts were accepted after revision.

Can staying overnight affect traveler satisfaction? Evidence from a gambling destination

ABSTRACT Given that satisfied travelers tend to become repeat customers and spread positive word of mouth, destinations must induce travel satisfaction to increase their income. Travelers may face increasing physical and mental challenges during their travel that can undermine their travel experience, and staying overnight in a destination can improve the satisfaction of travelers by refreshing their energy. Given its important role in increasing the revenues of certain destinations such as Macau, staying overnight is strongly encouraged among travelers. However, the relationship between staying overnight and traveler satisfaction in the casino context has never been studied. By conducting a survey among 17,742 travelers in Macau, this study explores the moderating effect of staying overnight on the relationship between travel experience and satisfaction. The results indicate that staying overnight strengthens (reduces) the negative effect of visiting frequency (outside-Macau casino experience). However, staying overnight has no moderating effect on such relationship when the negative influence of first-time visit is considered. These findings also provide implications for tourism researchers and practitioners.