Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Constructing Flight Itineraries with Machine Le...
Search
techsessions
February 14, 2018
Technology
0
6.6k
Constructing Flight Itineraries with Machine Learning
Dima Karamshuk, Senior Data Scientist, Skyscanner
techsessions
February 14, 2018
Tweet
Share
More Decks by techsessions
See All by techsessions
Building Multilingual Recommendations Systems for BBC News
techsessions
1
6.6k
Bayesian Online Change-Point Detection at Scale
techsessions
2
7.3k
Modeling the Importance of Flight Partners at Skyscanner
techsessions
0
6.6k
Automating Fraud Detection - Continuous Model Deployment
techsessions
0
6.7k
Natural Language Processing in Media: Challenges and Opportunities
techsessions
0
14k
The Impact of Automation at Scale
techsessions
0
8k
Machine Learning at Zopa
techsessions
0
8.1k
The Inner Workings of Monzo’s Help Search Algorithm
techsessions
3
14k
Modern Techniques for Dimensional Reduction
techsessions
1
14k
Other Decks in Technology
See All in Technology
AWSで始める実践Dagster入門
kitagawaz
1
620
DDD集約とサービスコンテキスト境界との関係性
pandayumi
3
280
BPaaSにおける人と協働する前提のAIエージェント-AWS登壇資料
kentarofujii
0
140
会社紹介資料 / Sansan Company Profile
sansan33
PRO
6
380k
サラリーマンの小遣いで作るtoCサービス - Cloudflare Workersでスケールする開発戦略
shinaps
2
450
Terraformで構築する セルフサービス型データプラットフォーム / terraform-self-service-data-platform
pei0804
1
180
Webアプリケーションにオブザーバビリティを実装するRust入門ガイド
nwiizo
7
830
Rustから学ぶ 非同期処理の仕組み
skanehira
1
140
MCPで変わる Amebaデザインシステム「Spindle」の開発
spindle
PRO
3
3.3k
Django's GeneratedField by example - DjangoCon US 2025
pauloxnet
0
150
[ JAWS-UG 東京 CommunityBuilders Night #2 ]SlackとAmazon Q Developerで 運用効率化を模索する
sh_fk2
3
430
【初心者向け】ローカルLLMの色々な動かし方まとめ
aratako
7
3.5k
Featured
See All Featured
The Myth of the Modular Monolith - Day 2 Keynote - Rails World 2024
eileencodes
26
3k
Product Roadmaps are Hard
iamctodd
PRO
54
11k
Measuring & Analyzing Core Web Vitals
bluesmoon
9
580
VelocityConf: Rendering Performance Case Studies
addyosmani
332
24k
Why You Should Never Use an ORM
jnunemaker
PRO
59
9.5k
Building Adaptive Systems
keathley
43
2.7k
Git: the NoSQL Database
bkeepers
PRO
431
66k
Principles of Awesome APIs and How to Build Them.
keavy
126
17k
Keith and Marios Guide to Fast Websites
keithpitt
411
22k
Fantastic passwords and where to find them - at NoRuKo
philnash
52
3.4k
The Illustrated Children's Guide to Kubernetes
chrisshort
48
50k
Unsuck your backbone
ammeep
671
58k
Transcript
Constructing Itineraries with Machine Learning @karamshuk
Skyscanner Mashups Virgin Delta Potentially cheaper itineraries in more than
half of all search results
Example
Main Difficulty with Mashups • Combinatorial explosion of variants •
Cost of constructing mashups is proportional to the volume of variants
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
Supervised Learning Classify whether for a query Q a combination
of partners (X and Y) is going to be in Top-10 cheapest
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
Temporal Validity Performance of the model stales, hence needs to
be updated regularly
Production Pipeline Work in progress…
Temporal Stability (Origin, Destination, Provider) rules We need a mechanism
to control temporal stability of the model
Join our Team!
[email protected]
on Twitter: @karamshuk @SkyscannerEng