onsdag 18 juli 2018

Tools that statisticians need to work as a Data Scientist

RoyalStatSoc Publicerades den 18 juli 2018

Aimee Gott - Mango Solutions What does a statistician need to know about machine learning? The rise of data science has changed the way in which we, as statisticians, need to think and work. New techniques are augmenting those of traditional statistics but they come with challenges for the statistician. From a new vocabulary to an alternative way of model development, there is an increasing need for us to move away from being the traditional focused statistician to the multi-faceted data scientist with a broader look on analytic approaches. But what do we really need to know to survive in this new world of analytics? In this talk we will delve into some of these ideas, highlighting some of the key machine learning algorithms and approaches to modelling so we can adapt to this new world.

Matthew Upson - Government Digital Service All the other things a data scientist needs to know Data science is a mixed discipline drawing from a number of fields. Whilst a good understanding of statistics and machine learning is critical for the role, equally, data scientists tend to be adept at a range of other skills, many coming from the software engineering world. This presentation explores ‘the other things’ data scientists need to know that are not generally covered in a typical statistical education: from working in an agile team to continuous integration, and (almost) everything in between.

Richard Pugh - Mango Solutions Data Science: Engaging with the Business The "Data Science" movement has led to a rise in popularity of proactive analytics to inform better decision making. However, this shift from analytics as a "reactive" study to a more strategic practice has led to the demand for modellers with a broader skillset. Companies who are striving to be "data driven" are desperate to hire analytic "unicorns" who can code like a software developer and analyse data using an increasingly range of analytic approaches. Perhaps beyond the "technical" skills demanded of a Data Scientist, there is a need for analytic teams to be able to interact with business functions in a more collaborative manner. This presentation will look at the demands on the modern data scientist from a "business" perspective, including the use of language, the exploration of analytic opportunities, the discuss of data science "success" and storytelling with data.