Kaggle Master Workshop #1: Telstra Network Disruptions Competition

The long-awaited Bath ML Meetup's Kaggle session is finally here! Dr Yifan Xie will run a few Kaggle workshops for Bath ML Meetup in the next couple months. Our first workshop will be on the Telstra Network Disruptions competition (https://www.kaggle.com/c/telstra-recruiting-network).

Dr Yifan Xie started his data science journey with his doctorate research about semantic computing in the aerospace, and gradually developed his machine learning knowledge with a combination of MOOC study and industrial projects. In 2017, Yifan co-founded Arion.ai - a data science consultancy aimed to serve the data science needs of start-ups, SMEs and enterprise level operations.

Yifan started competing on Kaggle, Google’s data science competition platform, since 2015. Currently he ranked 524 out of 65000+ competitors and achieve his Kaggle Master status in August 2017. Yifan also co-moderate the KaggleNoobs (https://kagglenoobs.herokuapp.com/) slack community - a social place for Kagglers and data science enthusiasts to exchange on anything from predictive modeling techniques to hardware configuration for machine learning.

A selection of Yifan’s Kaggle Competition Records (Ranking/Number of Participants)::
Quora Question Pairs (2017) - Top 1% (12th/3307) - Gold Medal
Santander Product Recommendation (2016) - Top 3%(38th/1785) - Silver Medal
Bag of Words Meets Bags of Popcorn Sentiment Analysis- Top 4%(18th/578)
Instacart Market Basket Analysis (2017) - Top 4% (79th of 2623) - Silver Medal
Grupo Bimbo Inventory Demand (2016) - Top 8%(150th/1969) - Bronze Medal

Schedule:

October 4th - Kaggle Newbie Workshop by BMLM organisers

October 18th - Kaggle Master Workshop #1

6:30pm - Doors open, refreshments

7:00pm - Introduction
• Introduction to the Telstra Network Disruptions competition
• An overview on the competition purpose & evaluation metric
• An overview on the dataset
• An overview on Yifan’s kaggle pipeline

7:30pm - Technical Workshop
• Research on the competition, EDA, model building, etc
• Submissions of our own work

You will need:

• Feature engineering tools such as numpy, pandas in Python, data.table, dyplr in R (anything you are comfortable with)

• Dataset (https://www.kaggle.com/c/telstra-recruiting-network/data) downloaded

Please invite your colleagues and friends so you can work in teams, or please feel free to come by yourself to work individually and/or collaborate with others. Hope to see many of you there at our first exciting Kaggle workshop! You can also bring your own Kaggle problems to share with Yifan.