Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Competing on Kaggle

Presentation for July Data Scientist Meetup in Bristol, UK
by

Yifan Xie

on 8 November 2016

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Competing on Kaggle

Thank you
Competing on Kaggle
- learning, fun and overfit if you dare
Something about me...
Background in computer science (BSc, MSc, EngD)
Work as Project Manager in Airbus
Know a bit of coding (VB, Python, R...)
Competing on Kaggle since 2015
Yifan Xie
a platform for data science competitions
Sponsored by some top organisations
500,000+ competitors

Problem Description
Data
Evaluation Metric (Cost Function)
Exchange & Sharing
Compete as Individual or Team
Model building & evaluation
Result Validation & Publication
Awards for Winners
Global Ranking & Achievements
Solution Sharing
Problem Definition
Competition
Post Competition
Problem Description
Data
Evaluation
Exchange & Sharing
Model building & Evaluation
Public Leader Board
Global Ranking & Achievements
Learning from the Best
Extended Community
Data Science - We never had it so good!
Theory
Tools
Playground
Have you overfitted!?
A painful Experience
Public Leader Board
Private Leader Board
Final Result (Private Leader Board)
?
---- Winning data science competitions, Owen Zhang
Overfitting
Learning Points
Robust Validation is EVERYTHING!
Personal Experience
It is very addictive
It is very time-consuming
https://www.kaggle.com/khyh00/introducing-kaggle-scripts/xkcd-style-test/run/60615
The Kaggle Community is the BEST community!
It is tons of fun!
Formula one of data science
BEST Learning Experience - EVER
Summary
Data Science that pushes predictive modelling to the limit
Kaggle is:
Best learning experience - build, fail and (hopefully!) succeed QUICKLY
Best Community - Sharing & Friendships
No model is correct, some are useful
Full transcript