Negative Airbnb reviews: an aspect-based sentiment analysis approach

Abstract

Purpose

The current paper aims at exploring negative aspects in reviews about Airbnb listings in Athens, Greece.

Design/methodology/approach

The aspect-based sentiment approach (ABSA), a subset of sentiment analysis, is used. The study analyzed 8,200 reviews, which had at least one negative aspect. Based on dependency parsing, noun phrases were extracted, and the underlying grammar relationships were used to identify aspect and sentiment terms.

Findings

The extracted aspect terms were classified into three broad categories, i.e. the location, the amenities and the host. To each of them the associated sentiment was assigned. Based on the results, Airbnb properties could focus on certain aspects related to negative sentiments in order to minimize negative reviews and increase customer satisfaction.

Originality/value

The study employs the ABSA, which offers more advantages in order to identify multiple conflicting sentiments in Airbnb comments, which is the limitation of the traditional sentiment analysis method.

Citation

Vassilikopoulou, A., Kamenidou, I. and Priporas, C.-V. (2022), "Negative Airbnb reviews: an aspect-based sentiment analysis approach", EuroMed Journal of Business, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/EMJB-03-2022-0052