Analyzing passengers’ emotions following flight delays- a 2011–2019 case study on SKYTRAX comments
Introduction
With the rapid development of new media and information technology, people tend to express their feelings and exchange opinions on social media and public social platforms. Social platforms have become a huge data source. Deep analysis of text data on social platforms and mining meaningful information for enterprises has become a research hotspot in the era of big data.
According to the 42nd Statistical Report on the Development of Internet in China issued by the China Internet Information Center on August 20, as of June 2018, the number of Internet users in China reached 802 million, with an Internet penetration rate of 57.7% (CNNIC, 2018). Even though traditional news and information software is still mainstream, it is undeniable that the rise of social media provides users with a new choice. Social media becomes an important way to spread hot news. According to the Research Report on user behavior of Chinese social application published by China Internet Information Center, social application has become an important channel for netizens to obtain news information and disseminate hot events (CNNIC, 2018). Pak and Paroubek (2010) argue that micro blogs have become a rich data source for opinion mining and sentiment analysis. Social applications, as platforms and intermediaries, contain abundant text data. User-generated content constitute a corpus for text analysis.
In recent years, China's aviation industry has developed rapidly, but there are still unavoidable problems behind it. According to the National Civil Aviation Flight Operational Efficiency Rathan et al., 2018 issued by the Civil Aviation Administration of China, the normal rate of civil aviation flights in 2017 is 71.67%, while the total number of flights reaches 5.320 million (CAAC, 2018a). In recent years, with the continuous increase in transport volume, the on-time rate of flights has always fluctuated around 75% in Fig. 1. Through research, Xu and Li (2016) report that flight delays are the main cause of most passengers' dissatisfaction.
Because of the openness of social platforms, social application platforms have become fertile ground for users to comment. Whether they are anecdotes, life stories, news reports, or experience of products and services, users can share them on social media platforms. When products or services do not meet customer expectations, customers are willing to express personal opinions on social platforms (Geho et al., 2010). And on the micro-blog platform, airlines and passengers have a high enthusiasm for interaction. Measured by the response rate, the aviation industry ranks second among all industries. Therefore, the importance of customer opinions for airlines is evident. Aviation service also has product attributes. Flight delay is obviously not expected by the customers. The gap between expectation and reality inevitably breeds discontent among passengers. Therefore, passengers tend to publish their personal feelings on the internet platform. Analyzing the factors affecting such discontent will help to improve the quality of service of enterprises.
With the rapid development of the aviation industry, the scale of air passengers is huge and thus the scale of people who experienced flight delays. Therefore, this paper analyzes the public opinion of flight delays. The user comments on SKYTRAX are collected for study.
Simple sentiment classification can only get the customers' sentiment inclination, which has limited practical value for industry. The objective of this paper is to apply text mining technology to automatically and efficiently access the information on flight delay text comments, applying a lexicon-based sentiment analysis tools to make sentiment analysis, and uses co-occurrence analysis to identify passengers' concerns on different aspects of service after flight delay in the aviation industry.
The rest of this paper is organized as follows: Section 2 provides a literature review. Section 3 shows data collection and preprocessing. Section 4 provides analysis of flight delay, and lexicon-based sentiment analysis dictionary to classify user comments. Section 5 concludes and provides some future research directions.
Section snippets
Text sentiment analysis
We use text mining to analyze public opinion for flight delays. Text mining, also known as text data mining or knowledge discovery based on text database, usually refers to the process of extracting valuable patterns or knowledge from unstructured text documents (Tan, 1999).
Previous work focuses on the analysis of positive and negative sentiment or the analysis of positive, negative and neutral emotional texts. However, it is difficult to describe complex inner world of human beings based on
Data collection
Since airlines do not provide dedicated websites for passenger reviews, this paper collected reviews on SKYTRAX between September 2011 and March 2019. SKYTRAX is an unofficial international air transport rating organization that provides the world's only airline quality ranking business. Passengers can find the corresponding airline or airport on the organization's official website, and publish a personal experience evaluation for their service. SKYTRAX summarizes the user's comments and
Lexicon-based sentiment analysis tools
This paper uses Vader and Pattern lexicon to analyze the sentiment of reviews (CAAC, 2018b). Vader is a rule-based framework for social media sentiment analysis. Each word's sentiment tendency is marked in the range of [-4,4], where 4 represents very positive, −4 represents very negative and 0 represents neutral. Using Vader to conduct sentiment analysis on a piece of text will output four values, marked as positive, negative, neutral and compound scores. First three values represent the
Conclusion
With the development of information technology, the channels of public opinion expression are expanding. For various industries, users' personal opinions on the network platform are increasing gradually. The maturity of text mining technology enables researchers or enterprises to automatically and efficiently access the information in text comments.
This paper uses crawler technology to obtain user reviews of several airlines on SKYTRAX and preprocesses the data. The lexicon-based sentiment
CRediT authorship contribution statement
Cen Song: Conceptualization, Writing - original draft, Funding acquisition. Jingquan Guo: Methodology, Software, Formal analysis, Data curation. Jun Zhuang: Writing - review & editing, Supervision, Project administration.
Acknowledgments
This research first two authors was partially supported by the National Natural Science Foundation of China under award number 71901218 and Beijing Social Science Foundation under award number 16GLC076.
References (28)
- et al.
Analysis of the perceived value of online tourism reviews Influence of readability and reviewer characteristics
Tourism Management Journal
(2016) - et al.
What makes tourists feel negatively about tourism destinations? application of hybrid text mining methodology to smart destination management
Journal of technological Forecasting & Social Change
(2017) - et al.
A study on the impact of airline service delays on emotional reactions and customer behavior
J. Air Transport. Manag.
(2016) - et al.
Predicting the performance of online consumer reviews A sentiment mining approach to big data analytics
Decision Support Systems Journal
(2016) - et al.
An empirical examination of a model of perceived service quality and satisfaction
J. Retailing
(1996) - et al.
An analysis of Brazilian flight delays based on frequent patterns
Transportation Research Journal, E
(2016) - et al.
A sentiment analysis of air passengers of top ten u.s. based airlines
Atlantic Marketing Association Proceedings
(2017) - et al.
Linking perceived service quality and service loyalty: a multi‐dimensional perspective
Eur. J. Market.
(1999) Statistical bulletin on the development of civil aviation industry in 2017
(2018)VADER sentiment analysis
(2018)