Consumer sentiment analysis of heritage hotels’ online reviews, using text mining: A multi-language approach (A case study of Noor mahal palace, Karnal)

Abstract

Purpose: In business, online reviews have an economic impact on firm performance.The study is an analysis of the online reviews written by customers about Noor Mahal palace, a heritage hotel in karnal, Haryana. It aims to analyze reviews with polarity classification and to identify the overall sentiment of the collected reviews.Design/methodology/approach: A total of 1,000 online reviews about the hotel were collected from the TripAdvisor website and web scraping was done by using a beautiful soup package in Python. Orange data mining software was applied to pre- process and analyze the data. VADER sentiment analysis is used to generate the polarity and overall sentiment of the reviews. The study categorizes the emotions of reviews as positive, neutral, and negative.Findings: Results revealed that 99.9% or 999 out of 1000 responses have a compound sentiment of positive and only 0.01 or 1 out of 1000 responses have a compound sentiment of negative. Reviews that have received a rating of 5 or 4 are regarded as being favorable reviews, whereas reviews that have received a rating of 1 or 2 are regarded as being negative reviews.Originality/value: The study is a rare attempt to examine online customer reviews of Noor Mahal palace using a text mining approach and to identify the sentiment scores of the online reviews to know the satisfaction and dissatisfaction.