Constructing Flight Itineraries with Machine Learning

Dima Karamshuk, Senior Data Scientist, Skyscanner

Transcript

1. Constructing Itineraries with Machine Learning @karamshuk
2. Skyscanner Mashups Virgin Delta Potentially cheaper itineraries in more than half of all search results
3. Example
4. Main Difficulty with Mashups • Combinatorial explosion of variants • Cost of constructing mashups is proportional to the volume of variants
5. Start Small Virgin Delta Sum of One Ways • fly-in with one airline, return with another • less combinations, we can effectively sample the ground truth space
6. Supervised Learning Classify whether for a query Q a combination of partners (X and Y) is going to be in Top-10 cheapest
7. Results Coverage or Recall: How many of all possible cheap Mashups our method recalls Cost or Quote Requests: How much quote requests we require for that
8. Temporal Validity Performance of the model stales, hence needs to be updated regularly
9. Production Pipeline Work in progress…
10. Temporal Stability (Origin, Destination, Provider) rules We need a mechanism to control temporal stability of the model
11. Join our Team! Dima.Karamshuk@skyscanner.net on Twitter: @karamshuk @SkyscannerEng