Predicting airline customers' recommendations using qualitative and quantitative contents of online reviews

Customers generally give ratings and reviews for different services that they get online or offline. These reviews and ratings aspects are effectively helpful to both the company and customers to receive feedback and make the right decisions, respectively. However, the number of reviews and ratings can increase exponentially, bringing a new challenge for the company to manage and track. Under these circumstances, it will also be hard for the customer to make the right decision. In this work, we summarize text reviews and ratings given by passengers for different airlines. The objective of this research is to predict whether the recommendation made by the customer is positive or negative. Two types of features, namely, textual feature and explicit ratings, are extracted from the dataset and other attributes. We found the relationship between such sentiments and feelings expressed in online reviews and predictive consumer recommendation decisions. We have considered quantitative content with qualitative content of online reviews in predicting recommendation decisions, which shows the work's novelty. Additionally, the obtained results yield an essential contribution to the existing literature in terms of service evaluation, making managerial policies, and predictive consumer recommendations, etc. Moreover, we hope that this work would be helpful for practitioners who wish to utilize the technique to make the quick and essential hidden information by combining textual reviews and various service aspects ratings.

Automated summary

This research analyzes passenger reviews and ratings for airlines to predict customer recommendations. The study observes a relationship between sentiments expressed in reviews and predictive consumer decisions. The findings contribute to service evaluation, policy-making, and predictive consumer recommendations.