November's data meet-up.
Pizza, beer and networking 6.30pm
Talk 1 (19:00) - Jens Rasmussen - "Fish and Chips: The data journey of fish from the ocean to our plates."
Talk 2 (19:45) - Dr Eyad Elyan, RGU - "Learning from Imbalanced Datasets: Challenges and Opportunities with Real World Examples and Applications"
Talk 1 - Jens Rasmussen.
Jens works as data manager in Marine Scotland – a government directorate tasked with looking after Scottish Seas. Arriving in Scotland 20 years ago, he initially worked as a research scientist in marine ecology. 10 years ago, he formally took on the job as data manager, and have since worked on opening up data and information about the Scottish Seas, improving the way in which data are stored, organised and published. He also collaborates with colleagues across the UK and Internationally through membership of various executive, advisory and expert groups on marine data management. But he also still dives into code, ancient paper records, and the occasional trip out to sea to collect more data.
Talk Description: Have you ever tried to calculate the number of earth worms in your back garden using only a tea spoon? The task is not too dissimilar from what scientist need to do to quantify the number of fish we can catch to put on our plates, while ensuring there are enough fish to catch in following years. There is a long process of capturing data on fisheries activates, and fish stock abundances in Scottish waters and beyond. The base for much of this activity is at the Marine Laboratory in Aberdeen. This talk will take you through examples of technology used and data collected, to be able to provide scientific advice about the amount of fish we can catch while keeping Scottish Seas sustainable.
Talk 2 - Dr Eyad Elyan
Class-imbalanced datasets are common across different domains such as health, banking, security and others. With such datasets, the class of interest is significantly underrepresented in the dataset leading to biased results. This talk will present some of the latest work that has been developed at the School of Computing using advanced deep learning models, and discuss how it was applied to solve real-world scenarios including hand-written characters/digits recognition, and analysis of complex engineering drawings.
Dr. Elyan has 20 yeas of experience in Academia and Industry. He obtained his first degree in Computer Science in 1999 from Al Quds University (Palestine). In 2004, he was awarded an MSc in Software Engineering with distinction. In 2008 at the University of Bradford obtained his PhD for his work on 3D modelling, representation, and recognition of human faces from 3D images.
Currently, he is a Reader (Associate Professor) in Machine Learning at the School of Computing Science and Digital Media at Robert Gordon University and he is leading the Machine Learning and Vision Applications research group. His research in the area of machine learning and machine vision is primarily focused on creating novel methods to represent and extract knowledge from unstructured and complex documents and images and structured data and applying it to solve industrial problems.
Elyan is a Fellow member of the British Higher Education Academy. He served as a Program Committee member for several international conferences and as a reviewer for several international journals in the area of machine vision, machine learning and data analytics and he has published more than 50 articles in the are of Deep Convolutional Neural Networks, Generative Adversarial Neural Network and Ensemble Learning.
More information projects, research output and others can be found at www3.rgu.ac.uk
Thanks to MBN Solutions, Data Lab Scotland and Scotland IS for sponsoring this event.
Photo by Sime Basioli on Unsplash