Hotel Recommendation System

Hotel Recommendation System

The process of choosing a hotel for a vacation or business trip is becoming more and more complicated. An algorithm-based hotel recommendation service appears to be a highly demanded solution to this hotel industry problem. Our colleague from the Institute of Mathematics of the Siberian Branch of the Russian Academy of Sciences participated in a Kaggle contest for the best recommendation service algorithm for Expedia and took third place among more than 2,000 teams specializing in Data Science.
https://www.kaggle.com/c/expedia-hotel-recommendations/
Industry:
Travel
Region:
USA
Technologies:
Python
Volume:
<= 1 man-year

Problem

At the launch of the project, Expedia had no data for analytical modeling and forecasting.

Thus, one of the project objectives was to develop a data model taking into account multiple parameters, which would later allow doing high-quality analytics and providing customers with hotel recommendations that suited them best.

Solution

Expedia provided a selection of data on the customer behavior over the years, the search criteria, the booking offers viewed, and the hotels booked.

Based on the search criteria and some additional information about the customer, the algorithm generated a list of personalized hotel recommendations.

The algorithm took third place in the Kaggle contest.

We use cookies.