Lots of B, or a single A?
Get in touch today
0141 225 0130
Last week, I had the pleasure of attending my first ever Scotland Data Science and Technology Meetup, right here in our MBN office! For anyone who doesn’t know what this group is, it is open to anyone interested in current and emerging Data Science, Analytics and Technology trends. This group was started in order to build a community to discuss general insights into the market and share personal experiences of working and operating in this area. I’m just entering my fourth week here at MBN, but I have already jumped on the bandwagon of feeling the hype for these events, and there is good reason!
Our speaker for this event, Pawel Dlotko, is a lecturer in Maths at Swansea University, so we were honoured to have him in Scotland sharing his wisdom with us all! His unique way of analysing data was new to everyone, judging by some mesmerized faces in the audience! The session was about using shapes and patterns to analyse data of all different kinds. Pawel began to describe his theory by stating that there are various concepts and abstractions known to humanity. The first and one of the easiest is the concept of counting different objects. It gives rise to almost all existing mathematics. But, he asks, what about shape? Intuitively we all feel that we see similar shapes emerging in different contexts but quantifying them is more complicated than simply counting them. Pawel went on to explain that the field of topological data analysis is an attempt to provide several invariants that are commutable and robust to encapsulate the shape of data. They can be used as additional features of your data in statistical analysis, machine learning and neural works for data of very different dimensions and kinds, including point cloud, digital images, results of numerical simulations, correlation patterns and beyond.
Towards the end of the session, where the audience was lucky enough to see some real life examples of this theory being brought to life, we saw that visualisations like this have been used to determine the survival rate in cancer patients, the best places for real estate investors to put their efforts, and even to describe the Brexit campaign!
The Scotland Data Science and Technology events are only getting stronger, so why not be a part of it? If you’re still not convinced, you’ll be provided with pizza and wine, beer or an ice-cold glass of Irn Bru! What more could you want?
Thanks again to Pawel Dlotko, and if you are interested in learning more about his work, please take a look at his YouTube channel.
Author: Chloe Lobley