Background

In 2017, an airline was launching a new accommodation booking service. They had limited data on guests using this new service and wanted to explore how it could offer a more personalised experience to these customers. The two main approaches to recommender systems are Collaborative Filtering and Content-based approaches, which do not perform well where there is limited data.

Technology Overview

A real-time hybrid recommender that combines different techniques and exploits all the available information about users, such as:

User preferences to personalise recommendations
Group preferences to capture preferences of similarly grouped people
Data associated with items to apply content-based techniques
Contextual Information

Inputs: User preferences and contextual information.

Adapt Technology: Hybrid approach that blends elements of naïve collaborative filtering, content-based recommendation and contextual suggestion.

Outputs: Custom accommodation recommendations for guests using the service.

Benefits

A recommender that overcomes the cold start problems of content-based and collaborative filtering approaches
Deployed with the airline for evaluation
Does not require significant rating data
Generates personalised recommendations
Provides real-time and robust recommendations

Opportunity

License
Research collaboration

Next-Generation Recommender

Seeking