Methoden vergleichen

	Hospitality eWOM Analysis	TripAdvisor Review Sentiment Mining
Fachgebiet	Tourism	Tourism
Familie	Process / pipeline	Machine learning
Entstehungsjahr	2008	2008
Urheber	Stephen Litvin, Ronald Goldsmith & Bing Pan	Bo Pang & Lillian Lee (opinion mining); applied to hotel reviews by Zheng Xiang and colleagues
Typ	Framework and pipeline for measuring electronic word-of-mouth volume, valence and influence	Supervised/lexicon text-classification of review polarity and opinion
Wegweisende Quelle	Litvin, S. W., Goldsmith, R. E., & Pan, B. (2008). Electronic Word-of-Mouth in Hospitality and Tourism Management. Tourism Management, 29(3), 458-468.	Pang, B., & Lee, L. (2008). Opinion Mining and Sentiment Analysis. Foundations and Trends in Information Retrieval, 2(1-2), 1-135.
Aliasnamen	Electronic Word-of-Mouth Analysis, Online Review Influence Analysis, Hospitality Online WOM Measurement, Digital Word-of-Mouth Analytics	Online Hotel Review Sentiment Analysis, Travel Review Opinion Mining, Hospitality Review Polarity Classification, Tourism Review Sentiment Classification
Verwandt	4	4
Zusammenfassung	Hospitality eWOM analysis is the systematic study of electronic word-of-mouth, the consumer-generated reviews, ratings, posts and comments that travellers share online about hotels, restaurants, attractions and destinations. Litvin, Goldsmith and Pan (2008) set out the foundational framework, defining eWOM, classifying its channels by communication scope and level of interactivity, and explaining why it matters so much in hospitality and tourism, whose intangible products are difficult to evaluate before consumption and are therefore judged heavily through the experiences of others. The analysis treats this online word-of-mouth as data, measuring its volume, its valence (how positive or negative it is) and the experience dimensions it reveals, and links these to outcomes such as bookings, satisfaction and reputation. Xiang and colleagues (2015) showed how large-scale text analytics of guest-generated reviews can deconstruct the hotel experience and connect it to satisfaction.	TripAdvisor review sentiment mining applies opinion mining and sentiment analysis to the large volumes of online reviews that travellers write about hotels, restaurants and attractions on platforms such as TripAdvisor. Grounded in the opinion-mining methodology surveyed by Pang and Lee (2008), it uses lexicon-based or machine-learning text classifiers to determine whether a review, sentence or opinion is positive, negative or neutral, turning unstructured free text into structured sentiment data. Applied to hospitality, as demonstrated by Xiang and colleagues (2015) in their big-data analysis of hotel guest experience, the technique can go beyond an overall verdict to extract aspect-level sentiment, revealing how guests feel about specific facets like room, service, location, value and cleanliness. The result is a scalable way to read what thousands of guests are actually saying and to quantify the tone of a property's online reputation.