Language interpretation in travel guidance platform: Text mining and sentiment analysis of TripAdvisor reviews

Abstract

The opinions and feelings expressed by tourists in their reviews intuitively represent tourists' evaluation of travel destinations with distinct tones and strong emotions. Both consumers and product/service providers need help understanding and navigating the resulting information spaces, which are vast and dynamic. Traditional sentiment analysis is mostly based on statistics, which can analyze the sentiment of a large number of texts. However, it is difficult to classify the overall sentiment of a text, and the context-independent nature limits their representative power in a rich context, hurting performance in Natural Language Processing (NLP) tasks. This work proposes an aspect-based sentiment analysis model by extracting aspect-category and corresponding sentiment polarity from tourists' reviews, based on the Bidirectional Encoder Representation from Transformers (BERT) model. First, we design a text enhancement strategy which utilizes iterative translation across multiple languages, to generate a dataset of 4,000 reviews by extending a dataset of 2,000 online reviews on 1,000 tourist attractions. Then, the enhanced dataset is reorganized into 10 classifications by the Term Frequency-Inverse Document Frequency (TF-IDF) method. Finally, the aspect-based sentiment analysis is performed on the enhanced dataset, and the obtained sentiment polarity classification and prediction of the tourism review data make the expectations and appeals in tourists' language available. The experimental study generates generic and personalized recommendations for users based on the emotions in the language and helps merchants achieve more effective service and product upgrades.

Computerized Content Analysis of TripAdvisor Reviews: A Case Study of a Street Food Restaurant With One Michelin Star

“…Apart from the proper nouns of the restaurant name 'Jay Fai', the other top frequent words are 'food', 'wait', 'crab', 'hour', 'restaurant', 'place', 'omelet', 'reservation', 'time', 'good', 'make', 'eat', 'table', and 'try' respectively. The finding of the word 'food' as the most frequent word is in line with previous research studies [3], [6], [25], [28] as well as the positive qualifying adjective 'good' [25]. [26].…”

“…Significantly, the word 'price' was found in most of the related research studies in different places: Turkey [24], Thailand [19], [3], and Malaysia [28]. This finding shows the fact that customers are really concerned about prices.…”

“…The words show that prices go with the good quality of ingredients. The word 'good' as one of the most frequent words is in line with [25], and the 'price' which was found as one of the most frequent words indicating negative aspect [19], [3], [28], [24]. Moreover, food quality was mentioned as one of the key variables on customers' perception [7].…”

“…Work [28] explored tourists' comments in travel guidance platform and found four factors representing the high expectations of tourists: food, price, crowdedness, and hygiene. Food was found related to hygiene in the comments and the expressions were mostly negative.…”

Sentiment analysis of travel reviews based on deep learning and transformer

“…A mounting number of tourists are contributing travel reviews concerning service quality, price and their attitudes of a scenic spot, giving rise to a surge in available tourism data [2]. These reviews are significant since they provide valuable insights into sentiments and overall satisfaction with the scenic spot [3]. Currently, tourists will refer to reviews from others before deciding on a destination.…”