An Examination of Sentiment Analysis as a Tool for Gathering Visitor Insights from Online Review Sites for a Museum

Abstract

This paper examines the online reviews from two popular cites (TripAdvisor and Yelp) for a presidential library and museum. A sentiment analysis is performed using two natural language processing (NLP) techniques (VADER and RoBERTa) in Python on each review cite and compared to determine the extent to which the reviews are similar between the cites and the NLP techniques. In addition, the results of the sentiment analysis for adjectives are used to predict the ratings given by the reviewers for the library. The results indicate a high degree of correlation between the cites for both the nouns and adjectives that appeared in the reviews. However, there are mixed results for the cities and NLP techniques in sentiment scores and predictive accuracy. This underscores the importance of using multiple NLP techniques and review sites to evaluate visitor sentiment.

Department(s)

Business and Information Technology

Keywords and Phrases

Online reviews; sentiment analysis; visitor insights

International Standard Serial Number (ISSN)

1540-6997; 1049-5142

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2025 Taylor and Francis Group; Routledge, All rights reserved.

Publication Date

01 Jan 2025

Link to Full Text

Share

 
COinS