Identifying Suggestions in Airline-User Tweets Using Natural Language Processing and Machine Learning

Identifying Suggestions in Airline-User Tweets Using Natural Language Processing and Machine Learning

Rafael Jiménez, Vicente García, Karla Olmos-Sánchez, Alan Ponce, Jorge Rodas-Osollo
DOI: 10.4018/978-1-7998-4730-4.ch022
OnDemand:
(Individual Chapters)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Social networks have moved from online sites to interact with your friends to a platform where people, artists, brands, and even presidents interact with crowds of people daily. Airlines are some of the companies that use social networks such as Twitter to communicate with their clients through messages with offers, travel recommendations, videos of collaborations with YouTubers, and surveys. Among the many responses to airline tweets, there are users' suggestions on how to improve their services or processes. These recommendations are essential since the success of many companies is based on offering what the client wants or needs. A database of tweets was created using user tweets sent to airline accounts on Twitter between July 30 (2019) and August 8 (2019). Natural language processing techniques were used on the database to preprocess its data. The latest classification results using Naive Bayes show an accuracy of 72.44%.
Chapter Preview
Top

Introduction

Nowadays, the influence that social networks have on people’s lives has caused marketing companies to change their business model (Saravanakumar & SuganthaLakshmi, 2012), going from advertisements in print media, radio and television to modern digital media, seeking to increase the reach of potential customers to whom transmit their advertising messages.

New digital platforms promote customer interaction and reward them with greater exposure as it becomes viral. Greater exposure to other potential customers by way of followers sharing their content creates value for the brands who use social media and their products (Ahmed, 2017). Social networks have moved from online sites primarily created to interact with our friends to a platform where people, artists, brands, and even presidents interact daily with crowds of people.

Through social networks, companies like DeWalt are making available the ability to co-create new products based on suggestions from their consumers (DeWalt, 2010). It has helped them stay on top of the construction tool manufacturing companies and win several awards with their products (DeWalt, 2019). This kind of innovation is called customer-driven innovation (Penisi & Kim, 2003) and requires an organization focused on delivering products with quality and attributes that consumers can assess positively.

According to a study published by the company Hootsuite (Kemp, 2019), Twitter is the seventh most accessed website in the world. It is a site where 500 million tweets are sent daily (Stricker, 2017). In Mexico, Twitter has 7.22 million users (Kemp, 2019), a figure that is on the rise and has shown a 3.7% increase in the last quarter of 2018.

Airlines are some of the companies that use social networks, such as Twitter, to interact with their customers through tweets, seeking to promote their services, offers, travel recommendations for a specific time of year, videos of collaborations with YouTubers or other companies, and publishing surveys to know the opinion of their customers on a particular topic, such as where in the world they would like to spend a weekend. Twitter users have three options to interact with each other through tweets:

  • 1.

    Sharing a tweet to the user’s followers through a retweet.

  • 2.

    Tagging a tweet as “Like” when users like an airline tweet, and

  • 3.

    Responding to a tweet issued by the airline through a comment in which users express their opinion or suggestion about the services provided.

In some comments that users respond to tweets issued by airlines, they express suggestions on how to improve their services or processes. The suggestions are important since the success of many of the companies is based on giving the customer what they want or need (Brun & Hagege, 2013; García, Florencia-Juárez, Sánchez-Solís, Rivera, & Contreras-Masse, 2019) and these needs may change over time.

Complete Chapter List

Search this Book:
Reset