Please log in to watch this conference skillscast.
Data science has emerged as "the sexiest job of the 21st century" - but can existing data science infrastructure and ecosystems live up to that title? Many data scientists use dynamically typed languages such as Python, or R. In the Haskell community, you know that functional programming and powerful type systems can deliver massively increased productivity by promoting composition. If this promise is to be projected onto serious, real-world data science, there is remains considerable work to be done for emerging competitive ecosystem.
Tom will briefly outline the data science process and the different kinds of workflows employed, depending on the questions that are asked and the volume and variety of data present. You will learn how, contrary to the common perception, there in fact is a substantial body of work existing in the Haskell open source ecosystem that can solve many of these tasks.
You will focus on two recent projects: combinator libraries predictive analytics and dashboard generation. The R language have mature projects in this area, caret and Shiny respectively. Implementing these projects in Haskell forces us to formalise the types of operations necessary; but this formalisation brings new opportunities in terms of presenting a constructor kit with building block that can be combined in ways that were difficult to imagine in a less formal approach.
YOU MAY ALSO LIKE:
- Lightning Talk: Why I No longer Use Haskell (SkillsCast recorded in October 2019)
- Leonardo De Marchi's Deep Learning Fundamentals (in London on 27th - 28th January 2020)
- Haskell eXchange 2020 (in London on 8th - 9th October 2020)
- Solandra Hands-On Tutorial & Emergent Behaviour In Insects (in London on 28th October 2019)
- Conditional Random Fields: Probabilistic Models for Segmenting and Labelling Sequence Data. (in London on 28th October 2019)
- Lightning Talk: Making and Testing Code Generators in Haskell (SkillsCast recorded in October 2019)
- GHC Runtime Linker by Example (SkillsCast recorded in October 2019)
Data Science in Haskell: Solutions and Challenges
Tom is a data science team leader building predictive analytics-based products, specialising in preference learning, visual analytics and marketing using Bayesian and deep learning methods, probabilistic and functional programming. He has a background as an experimental and computational neuroscientist - Tom obtained his PhD in Neuroscience from University College London by pouring slimy stuff on brain cells, which combined with a reaction-diffusion model allowed him to measure biophysical properties of synapses. As a Research Associate at Harvard Medical School, and then the Universities of Leicester and Nottingham, he built microscopes and Domain Specific Languages in Haskell to control them. Working at the intersection of experimental, theoretical and methodological neuroscience has given him a uniquely creative perspective on data science. As the Chief Data Science Officer at a creative social agency, he led a team team building a series of models for predicting and enhancing the impact an image will have in specific marketing contexts using Deep Learning model; attribution modelling from social media data, and a platform for delivering visual consumer advertising on social media. He is now working on an open source data science stack in Haskell.