Exploring Data on Airline Prices From TripAdvisor

Posted on Dec 15, 2017
The skills the author demoed here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.

Introduction

Data shows the airline industry is one of the biggest contributors to US economy. In 2016, the airline industry profits have reached $36 billion, and by the end of 2017, it was estimated that net profit will reach $30 billion. While there has been a lot data in recent years showcasing the rise in airline profits contributed by higher flight prices, identifying the causes for flight prices have still been pretty elusive. For frequent flyers, understanding how airlines structure their prices can allow them to use cost saving strategies to purchase flight tickets.

Breakdown Questions

  1. Time Series of Flight Prices : How does flight prices change by destinations over a one month period? What is the trend for flight prices for different airlines?
  2.  Flight Price Distribution : What are the flight price distributions by airline ? by day?

Data

To understand more about what factors contribute to flight prices, data was scraped from the website TripAdvisor which collects many information such as different airline flight prices, flight dates,  arrival and departure times, flight duration, flight scores,  and flight customer-written reviews of flights. A typical page is as follows:

Exploring Data on Airline Prices From TripAdvisor

A python script which uses Selenium defines a web scraping spider that scrapes information from the website and convert it into an analyzable format. Selenium was chosen as the package to scrape TripAdvisor rather than the faster package Scrapy because the website was a dynamic page. This meant that the web page displays  flight information which changes after a certain time frame. In addition, Selenium could handle information in click-on popups better. The code could be  found on Github .

Data was gathered for one way trips departing from New City to the top destinations in United States: Los Angeles, Honolulu, San Francisco,  Miami, and Las Vegas during a one month period, December to January. Around 3,000 pages were scraped to produce 47805 observations and 9 features.  The features I included in the scrape were:

  • airline name
  • flight cost
  • departure time
  • arrival time
  • flight duration
  • flight flyscore
  • airline rating
  • date of departure
  • review comments

One thing to note is that on some flight entries, there were missing flight flyscores which meant that these features would be replaced by 'NAs'.

Time Series Destinations

Exploring Data on Airline Prices From TripAdvisor

As seen from the above time series, travelling to Honolulu is almost double the price than any other destination throughout the one month period. This makes intuitive sense as Honolulu is one of the most popular tourist place in United States and fewer US airlines travel from other cities to Honolulu. As a result of the relative larger demand to supply of airline flights to Honolulu, flight prices are much higher for Honolulu than other destinations in United States.

Another interesting feature to note is that flight prices have a decreasing trend from the respective time frame between middle of December to middle of January. This trend is partly due to the fact that there is a large demand of flight travelers before Christmas; hence the more expensive flight prices. After Christmas and New Year, the demand for flights decrease and thus the decreasing trend of flight prices.

Time Series Airlines

Exploring Data on Airline Prices From TripAdvisor

When looking at time series of flight prices broken down by different airlines, we can see that the average prices exhibit very high volatility in a day to day basis. However, we can see that large airline corporations such as United, American, and Jetblue  have flight fares which are almost twice as expensive compared to smaller airline corporations such as Frontier and Sun Country for .

If we investigate fleet size, number of destinations, and daily departures between the bigger and smaller airlines, we find that the bigger airlines have 10 times the number of airplanes, about 5 times the number of travel destinations, and 200 times the number of daily departures. This means that larger airlines  dominate a larger portion of the airline industry and capture more of the consumer base; allowing them to increase flight prices relative to smaller airlines.

Flight Price Distributions

From the distributions above, we can get a sense of the  ranges of flight prices between December and January as well as distribution broken down by different airlines. We can see that the distribution of flight prices are left skewed, indicating that the mean flight price is higher than the median flight price. This is explained by the fact that some airlines price their flights at really high values, driving the mean price up. For example, if we look at the distribution of United Airlines, we can see that their minimum flight price is on par with other airlines within the range of $100-$150. However, their maximum flight prices range of $1000 and more which drove their mean price up.

Flight Price Distribution by Date and Airline

We have observed a wide range of flight prices broken down by airlines in the aggregate time-frame between December to January. An interesting observation I found while scraping TripAdvisor was that airlines set different flight prices on a given day for the same flight. This observation could be visualized above in the distribution of flight prices by airlines and date. In order to understand why airlines set different prices on a single flight, I did further research.

In my research, I found that airlines will segment a certain amount of seats with different prices to target different types of flyers. There are two different types of flyers: business and leisure. For business flyers, prices are flexible for them because the company pays for their trip or they can afford the higher prices. On the other hand, leisure flyers look for relatively cheap flights. In order to maximize profits, airlines appeal to those who are willing to pay more as well as those who settle for lower cost. As a result, they would use booking data to find the right combination of number of seats and price brackets for the seats.

For example United Airlines is known to increase the number of seats in airplanes; up to 250 seats for a domestic flight and thus allowing them to set more high price brackets to cater more to business flyers. They even have a class within economy class called economy plus to increase the flight prices for those seats.

Conclusion

Despite the fact that there are a myriad of factors that impact  flight costs, flyers can develop cost saving strategies by understanding the structure of flight prices and identifying flight price trends. One strategy to obtaining low airline prices is to book early for big airlines so that there are still available lower price bracket seats. The other option  would be to book smaller airlines where generally prices are cheaper than bigger airlines but at the cost of worse flight experience.

This project only scratches the surface in understanding flight prices and the factors which impact them. Future work could include predicting flight prices based on these sets of features as well as other features taken from weather and economic data. More features would help to develop an accurate machine learning model such as random forest to predict flight prices.  To take it further, I'd like to employ natural language processing to see whether customer reviews and ratings have an effect on flight prices and what features do customers really want in a flight experience. The insights from that study could lead to better improvement for flight experience and also maintain growth of profits for airlines.

 

About Author

Hans Lau

Hans graduated from University of Illinois Urbana-Champaign with a major in Bioengineering and a minor in Computer Science. He is passionate in the applications of data science especially in medical technology and biological systems. In his downtime, he...
View all posts by Hans Lau >

Related Articles

Leave a Comment

No comments found.

View Posts by Categories


Our Recent Popular Posts


View Posts by Tags

#python #trainwithnycdsa 2019 2020 Revenue 3-points agriculture air quality airbnb airline alcohol Alex Baransky algorithm alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus ames dataset ames housing dataset apartment rent API Application artist aws bank loans beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep boston safety Bundles cake recipe California Cancer Research capstone car price Career Career Day citibike classic cars classpass clustering Coding Course Demo Course Report covid 19 credit credit card crime frequency crops D3.js data data analysis Data Analyst data analytics data for tripadvisor reviews data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization database Deep Learning Demo Day Discount disney dplyr drug data e-commerce economy employee employee burnout employer networking environment feature engineering Finance Financial Data Science fitness studio Flask flight delay gbm Get Hired ggplot2 googleVis H20 Hadoop hallmark holiday movie happiness healthcare frauds higgs boson Hiring hiring partner events Hiring Partners hotels housing housing data housing predictions housing price hy-vee Income Industry Experts Injuries Instructor Blog Instructor Interview insurance italki Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter las vegas airport lasso regression Lead Data Scienctist Lead Data Scientist leaflet league linear regression Logistic Regression machine learning Maps market matplotlib Medical Research Meet the team meetup methal health miami beach movie music Napoli NBA netflix Networking neural network Neural networks New Courses NHL nlp NYC NYC Data Science nyc data science academy NYC Open Data nyc property NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time performance phoenix pollutants Portfolio Development precision measurement prediction Prework Programming public safety PwC python Python Data Analysis python machine learning python scrapy python web scraping python webscraping Python Workshop R R Data Analysis R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn seafood type Selenium sentiment analysis sentiment classification Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau teachers team team performance TensorFlow Testimonial tf-idf Top Data Science Bootcamp Top manufacturing companies Transfers tweets twitter videos visualization wallstreet wallstreetbets web scraping Weekend Course What to expect whiskey whiskeyadvocate wildfire word cloud word2vec XGBoost yelp youtube trending ZORI