Nic Crane, Data Scientist

The agenda for EARL London has just been released, and with three streams running in parallel, I always like to pick which talks I’ll be going to in advance, to make sure I get the most out of the conference. Here are my top picks so far:

Derek Norton, Microsoft: “SAS to R: How to, and is it enough?” – Day 1, 11am, Stream 3.
I’m a strong supporter of the promotion of free open-source software, so SAS to R is a topic that has instant appeal to me. Cards on the table – I’m co-presenting on a similar theme immediately afterwards, and so I’m looking forward to see the common themes that both Mango and Microsoft have identified on helping clients make the move to R.

Dr Troy Hernandez, IBM: “Using Twitter sentiment to predict stock price” – Day 1, 12pm, Stream 2.
With an undergraduate degree in psychology and a fascination with behavioural economics, I’m always interested in anything that combines data science and human behaviour. Sentiment analysis always seems to make for interesting results with some unexpected findings, so I’m looking forward to this talk.

Grace Meyer, Mango Solutions: “Too good for your own good: Shiny prototypes out of control” – Day 1, 3.30pm, Stream 2.
I’ve tried to avoid picking Mango talks so far. However, I had to pick this one as it has me really intrigued. I’m a huge fan of Shiny as a communication tool – I love how it can be used to communicate complex analyses to non-technical stakeholders in an engaging way. I don’t know anything about Grace’s talk yet, but it sounds like we’ll be getting to hear a bit more of a balanced approach to the prevalence of Shiny proof-of-concepts which seem to be sneaking into production more and more.

Dr Tim Paulden, ATASS Sports: “We ‘R’ young, we run green: Transforming cities through youth-driven data science” – Day 1, 3.30pm, Stream 3.
I met Tim last year at EARL London, and really enjoyed his talk on football modelling, as well as his ability to explain Benford’s Law in an intuitive way, even after I’d had a few beers! He’s a great presenter, and I look forward to his talk this year on youth-driven data science.

Nigel Carpenter, RSA: “5 tips to transform your analysts into high performing data scientists with R” – Day 2, 11.30am, Stream 3.
I mentioned before that I’m interested in SAS to R conversions, and an important part of this process is getting the existing developers on board and excited about R. I find all things data science fascinating, but appreciate that programming can be a dry topic, so I love to hear about how we engage people in a way that benefits both the individual and the business.

Annabel St John – Lyle, reed.co.uk: “Forecasting metrics with Facebook’s prophet package” – Day 2, 1.30pm, Stream 3.
Facebook recently made their in-house forecasting tool open-source, so I’m looking forward to hearing more about it, how reed.co.uk have been using it themselves and how they’ve found it compared to other forecasting methods.

Tom Wagstaff, Crisis: “How many property guardians are there in the UK? Scraping the web with R” – Day 2, 4.00pm, Stream 3.
This talk first struck me as one to take a look at as I’ll be teaching a workshop on web scraping on the Tuesday before the conference. That said, I love having interesting examples I can cite of how people are using the rvest package, and this looks like a genuinely fascinating talk, addressing the challenge of collecting data when the sources are limited.

Given the number of other topics that have caught my eye, this list will likely change before EARL London in September, and there’s a lot of sessions that I’m not totally decided for yet. What are your current favourites so far? Tell me on Twitter!