18th AIAI 2022, 17 - 20 June 2022, Greece

Deep Recurrent Neural Networks for OYO Hotels Recommendation

Anshul Rankawat, Rahul Kumar, Arun Kumar


  Recommendation Systems at OYO solve a complex personalization problem with scale and sophistication. The authors have focused on the development of the best-in-class recommendation system in the hospitality industry using a deep learning based model. The objective of the work is to develop a recommendation model which uses sequences of user interactions with contents derived from user interactions. The hybrid model described in the paper is a deep recurrent neural network based architecture split into two components: first, an embedding generation model and then a deep prediction and ranking model. The models have shown significant performance improvement both online and offline over existing collaborative filtering based models across geographies irrespective of traffic density and hotel supply density. The success of the deep learning based hybrid recommendation model at OYO across different geographies indicates immense potential of such recommender systems in industries such as travel, hospitality etc.  

*** Title, author list and abstract as seen in the Camera-Ready version of the paper that was provided to Conference Committee. Small changes that may have occurred during processing by Springer may not appear in this window.