David Azcona is originally from Pamplona, Spain. He completed his PhD under Prof Alan Smeaton’s supervision in Machine Learning applied to Computer Science Education. He applied supervised approaches to help students that are falling behind in programming courses at Dublin City University’s School of Computing. He predicted their performance based on their progress and recommended material and code snippets for their learning. Afterwards, as a postdoctoral researcher he worked on various projects from memorability, placing first at their annual workshop, to detecting knee injuries using MRI images and Computer Vision.
‘I joined fashion online retailer Zalando as an Applied Scientist in 2020. I first started working on Personalisation by recommending brands to our customers that are relevant to their personal style by learning preferences and serving them to their millions of customers. Traffic in ecommerce generally follows a 80 – 20 rule, around 80% of the traffic goes to 20% of the content so these recommender systems are typically biased towards popular rankings. Brand recommendations are even more skewed so I derived an approach with Prof Smeaton, to circumvent this problem and recommend new brands to interested users.
Afterwards, I joined Pricing and worked on matching products between Zalando’s assortment and their competitors. We trained Machine Learning models using multimodal signals including images and text to generate candidates to be reviewed by human agents before applying discounts to our products.
‘Recently, I joined Etsy, a global marketplace for buyers and sellers where you can find special items, from unique handcrafted pieces to vintage treasures. I work detecting content that violates our policies at scale in near real-time using Machine Learning.
‘During my PhD, I had the opportunity to spend a year at Arizona State University as a Fulbright fellow researching how their large student body learn online and in person. This was possible thanks to Insight’s connections and DCU’s partnership with ASU. Over there, I also leveraged Machine Learning to detect wildfires before they become catastrophic with some friends at the Microsoft Imagine Cup, see here.
‘The Insight Centre allowed me and other researchers to complete my PhD at the same time as other students at the school, university and other institutions in Ireland. This support network is crucial when you are going through this process as it allows you to ask questions, get feedback, attend and present at internal and external events and conferences. After my PhD, I came across many people, specially at Zalando, that have attended Insight as PhD researchers. Insight is a well-known and recognised institution that employers and interviewers recognise for their groundbreaking work in Machine Learning applied to many areas such as vision, lifelogging or healthcare.’