Data Science for Travel & Logistics at Trainline

Data Science Festival - London
Data Science Festival - London
Public group

Location visible to members


Join Data Science Festival - London in partnership with Trainline, TFL & this month for 3 great presentations.

Note: Please use your full real names where signing up, otherwise you will not be able to gain access to the venue. People registered with only their first names will be removed from the guest list and replaced from the waiting list.


6.00pm doors open[masked]:05pm - Talk - John Telford & Marco Rossetti
7:05 - 7:25 pm - Talk - Ben Cooper & Emma Beynon
7:25 - 8:05pm - Beer & Pizza
8:05 - 8.40pm - Talk - George Vafiadis & Hao Sky Zhou
8:40 - 9:00pm - Beer & Close

John Telford & Marco Rossetti: Data Science at Trainline for Smarter Journeys

Summary: Trainline is an online train tickets retailer for UK and Europe with 11 milion monthly visitors and more than 100,000 tickets sold every day. In this talk we will share how we are using Amazon Web Services for deploying data products using serverless architecture and providing analytics in real time. As a case study, we will present BusyBot, a special feature that helps you to locate train carriages most likely to have empty seats and more space, so that you can travel more comfortably.

Bio: John Telford, Head of Data Architecture at Trainline. Leading the adoption of Big Data technology at Trainline. Manages a team of Data Engineers and Database Administrators. Previously worked on Data Warehousing and Big Data at Channel 4. Computer Science degree from Brunel University. Twitter: @jtelford1

Bio: Marco Rossetti, Senior Data Scientist at Trainline. Leading personalisation initiatives, like providing context-aware personalised services, journey recommendations, and tailored travel options. Prior to joining Trainline, he worked at Mendeley, developing recommender systems for researchers. He has a PhD in Computer Science from University of Milan-Bicocca. Twitter: @ross85

Ben Cooper & Emma Beynon: London Underground - Improving asset performance through data science

Summary: London Underground holds a wealth of data on it’s service delivery assets, from when and how an asset has failed, to detailed minute-by-minute monitoring of asset condition. Our aim as data scientists is to leverage this data, along with external datasets where relevant, to deliver asset performance improvement, and ideally cost savings. We will present some examples of our work to date, and our plans for the future.

Bio: Ben Cooper, obtained a PhD in Particle Physics in 2006 from UCL and went on to spend nearly ten years as a physicist on the ATLAS experiment at the Large Hadron Collider, before joining London Underground as a data scientist in 2015.

Bio: Emma Beynon, obtained a PhD in Cosmology in 2012 from the University of Portsmouth and went on to publish numerous papers on Gravitational Lensing, before joining TfL as a data scientist in 2015.

George Vafiadis & Hao Sky Zhou: Methods in handling issues of massive data imbalance and correlation within target responses in machine learning.

Summary: ( is a leading price comparison site that hold massive amount of web-traffic and demographic data. Our aim as data scientists and data engineer are to identify opportunities to help potential customers save money across a diverse range of products, from car and home insurance; to energy; to personal finances, such as credit cards and loans through the use of these data. We will be sharing our newly developed methods in dealing with both data imbalance and correlation among target responses issues within the work. One can modify and implement these methods to solve similar problems during their data science projects.

Bio: George Vafiadis, is a senior software engineer responsible for the design and implementation of BigData Platforms at CompareTheMarket. He has MSc in Communications Systems and Signal Processing from University of Bristol and BEng in Information and Communication Systems Engineering from University of the Aegean.

Bio: Hao Sky Zhou, obtained a MSc Statistics from University of Warwick 2009 and a PhD in Econometrics and Statistics in 2012 from University of Leicester worked as senior data scientist at Sky Betting and Gaming before joining ( in 2016.