The persistent advancement in information technology has had a significant effect on the travel industry. The extensive public use of the internet has created situations that have been game changers in the field of traveling. The working differences between the travel agent and airlines is becoming vague in this modern age.
People who love to explore the beauty of this world or who are passionate about traveling often face problems regarding travel booking. With this, kayak.com saw an opportunity. Kayak is a company that aims to makes traveling better. It compares hundreds of sites simultaneously and finds the best and cheapest deals. Before Kayak, people had two options: visit a travel agent; search across multiple airline web sites. Kayak has given birth to new traveling generation that has positively disrupted travel experience as a whole. Kayak provides their service in over 30 countries and 20 languages.
Kayak, big data and predictive analytics
Kayak receives data from many different third parties and they have to spend time cleaning it. For example they might get a record about one hotel from Priceline.com and a different version from Travelocity. Kayak needs to rationalize that data, clean it up and create one unique record for every hotel. This is done with a lot of machine learning. They train machine learning models for how human experts would compare this data. The comparisons it’s not confident about, are kicked off to a human for further analysis. The technologies behind this are mostly Hadoop for data access, Python for dashboards, machine learning and Java for overall architecture.
Perspective travelers often feel overwhelmed seeing limitless options and find it difficult to decide what they actually want. Human emotions like these offer opportunities for Kayak to turn prescriptive analysis into predictive analytics. To this end, Kayak’s engineers came up with a strategy and applied it to A/B testing. Their users voted with their clicks – the features that users found useful were emphasized for production and the features that didn’t were eliminated.
Flight price prediction was a feature like this. Kayak’s analytics teams thought they could do something reasonably accurate with a machine learning model. They deployed it and their users used it, so they expanded it. Forecasting was done through a crowd sourcing model. The trips that were more popular for their users gave them better forecasting. As an example, during spring break, Kayak got better predictions for trips to the Caribbean. It’s an input for the users’ final decision, and Kayak tells their users how good they’ve been at that prediction in the past so they can decide if they buy now or wait until later.
Through predictive analytics, Kayak has set a new standard that fulfills the requirements of this changed industry. Their organic approach to customer behavior modelling is ever expanding. Kayak’s goal: build the best web and mobile site to deliver the best user experience and save people time and money. With their success it seems, everyone profits.
