Do you start your New Year with a Resolution? Many people do, and the results are usually not great. That is why I don’t set myself resolutions — I set a theme for the New Year. Popularised by YouTuber and podcaster CGP Grey, a yearly theme guides you to improve yourself throughout the year, rather than defining a rigid resolution that’s most likely not going to last beyond the first week of January. As an example, instead of promising to quit smoking or losing weight, you might define your year as the Year of Health and set these as areas…

Sports Analytics

Formula 1 races often feel predictable: Mercedes has won the past 5 championships and 7 of the first 9 GPs in the 2020 season. There is certainly no lack of Formula 1 predictions online — anyone from professional pundits to self-appointed armchair “experts” is often keen to share theirs. My goal was to take a more analytical look at the problem: could Formula 1 races be predicted by algorithms?

I have been working on predicting Formula 1 qualifying results for the past two years. From the start, my goal was to predict race results as well, but this turned out…

If you are a student planning a career in technology, you probably need no convincing over the value of internships. Internships allow students to gain professional experience, network with professionals, and earn money over the summer break. They can also be great fun, if you find a company that matches your values and style of working. As a two-time intern and a current graduate software engineer at Thought Machine, I wanted to share my experience to help you find the perfect internship! …

Ever since COVID-19 took the world by storm, schools have struggled to adapt to the new normal of remote teaching. While the crisis has caused great disruption, particularly for the most disadvantaged students, it also created opportunities. Technology can enable active learning methods, which have been found to improve learning outcomes over traditional lectures. Active and gamified learning has already found adoption in online education: language learning applications such as Duolingo do not ask you to watch a 50-minute lecture of Spanish grammar but teach using bite-sized lessons consisting of theory and exercises. Active learning can also work well when…

From seashells and animal skins to gold coins and paper notes, currency has functioned both as a store of value and a common unit for measuring the value of products and services across history. While currency has traditionally been tied to physical objects, most money in use today exists only as bits in data centres. These bits are not a new currency though: they measure wealth in dollars, euros, or yuan, and are convertible to their paper forms through banks and ATMs. Modern technology has also enabled a new type of money: cryptocurrencies like Bitcoin and Ethereum. Unlike the money…

This post is part 3 in my series of posts about explainable recommendations, based on my BSc dissertation. Part 1 introduces the concept of explainable recommendations, while part 2 covers my implementation of explainable movie recommendations using Python and React.js.

Recommender systems have seen increased usage over the past decade, and they’re considered an integral part of applications such as YouTube, Netflix and Amazon. While recommendations have helped users find the videos, movies, or items they like in a limitless ocean of options, they’ve not done so without problems. Recently some recommender systems have faced criticism for directing users to…

This post is part 2 in my series of posts about explainable recommendations, based on my BSc dissertation. Part 1 introduces the concept of explainable recommendations, while part 3 discusses the application of post-hoc explainability in data science.

The previous post in the series discussed why improving the explainability of recommender system matters. This is not a tutorial, but rather an overview of the approach I took in implementing a movie recommending service with explainable recommendations. If you wish to explore it further, you can read the full dissertation or dive into the front-end or back-end codebases (which are open…

This post is part 1 in my series of posts about explainable recommendations, based on my BSc dissertation. Part 2 shows the implementation of an explainable movie recommender system, while part 3 discusses the application of post-hoc explainability in data science.

Recommender systems help users discover new items, and they have seen increased use in various applications including video streaming, e-commerce, and social media. Recommender systems can be built using several classes of techniques, but here I will focus on collaborative filtering with latent factor models (LFM). Collaborative filtering methods produce recommendations based on the users’ previous ratings with the…