Shayan Doust

I am currently 20 years old studying Computer Science at the University of Liverpool. My interests can be pinpointed on algorithms and Data Science. I enjoy researching and learning about "the art" of Machine Learning and Artificial Intelligence, though I can safely say that this space extends to anything Mathematical.

I started to first use Debian back in 2013, although this became my daily-driver in 2018. Whether I am studying, socialising or tinkering, I always make time to contribute towards Debian; I definitely enjoy packaging. Most of my work within Debian is centred onto the Debian Med and R packaging teams, as they contain the packages I have the most experience with. Some of the packages within the Debian Med and R packaging teams are used quite frequently by me for my projects.

Accepted Talks:

Probing into the world of Data Science

We live in a world where we are relying more and more on data. The emergence (and progression) of computers have made data harvesting and storage much bigger both volumetrically and computationally. However, with the mass volume of data comes the question of “how can we extract any meaningful semantics out of this (usually unstructured) data to gain some sort of logical insight or sense, to perhaps build some sort of knowledge base?” This is where the umbrella term “Data Science” makes sense, jointly covering disciplines such as Data Analytics, Data Mining, Machine Learning, et cetera that are used to satisfy the (quite vague and abstract) aforementioned question.

This talk will cover a few fundamental points. Firstly, I would like to introduce the field of “Data Science” more in-depth including different exemplars and complications. Following this, I will talk about the common tools, techniques and languages used (such as libraries & programs) as well as an analysis of the current Data Science tools that exist in Debian; their ‘health’, Debian’s strong (and weak) points in this field and prospective packages that should be added.

This will be a pre-recorded talk. This description is very much the bare minimum, and indeed more fruitful content may be added.