Interactive AI PhD
University of Bristol, 2019-2023
Areas of Research:
- Active Learning
- Weakly-Supervised Learning
- Interactive Visualisation Software Development
Computer Science MEng
University of Bristol, 2015-2019
- Programming & Algorithms
- Applied Deep Learning
- Machine Learning
- Applied Data Science
- Computer Architecture
- Systems Security
- Computer Graphics
- 3D Modelling & Animation
- Software Product Engineering
University of Bristol | Student Ambassador
December 2016 - Present
I am also currently a student ambassador for my university. I participate in outreach programmes for school students such as Access to Bristol and University Summer Schools, as well as open days and campus tours. My role often requires me to solely lead a large group of students, requiring me to be independent and have strong communication and presentation skills.
University of Bristol | Senior Resident
September 2018 - September 2021
I am a live-in peer mentor within halls of residence. I provide advice to individual students and contribute to inclusive community-building events and activities in partnership with the student-led JCRs. It is also my responsibility to identify at an early stage those students who may be vulnerable and need referring to the Residential Life Team.
University of Bristol | International Office Intern
July 2017 - September 2017
During the summer break of my 2nd year, I was an intern for the University’s International Office. I was in charge of all correspondence, booking and leading of all tours for prospective international students around the university. I thoroughly enjoyed interacting with the visitors, answering questions about my experiences at Bristol and university life in general. With up to 3 tours a day, each with up to 5 family groups, it was critical to manage the bookings so that each group could receive the key information they required given that many had travelled long distanced and were on tight schedules.
C++ • Java • Matlab
Blender • Final Cut Pro X
Projects & Achievements
AstronomicAL: an interactive dashboard for visualisation, integration and classification of data with Active Learning
Developed as part of my PhD thesis
AstronomicAL is a human-in-the-loop interactive labelling and training dashboard that allows users to create reliable datasets and robust classifiers using active learning. The system enables users to visualise and integrate data from different sources and deal with incorrect or missing labels and imbalanced class sizes by using active learning to help the user focus on correcting the labels of a few key examples. Combining the use of the Panel, Bokeh, modAL and SciKit Learn packages, AstronomicAL enables researchers to take full advantage of the benefits of active learning: high accuracy models using just a fraction of the total data, without the requirement of being well versed in underlying libraries.
AstronomicAL resulted in a publication in the Journal for Open Source Software. More info can be found here.
Passive Information Extraction System (P.I.E.S)
Interactive AI Group Project - Developed for L.V Insurance
Based on the requirements of L.V, our group developed a backend question-answering system which assists the service desk staff in extracting relevant information from customer phone calls. The Passive Information Extraction System (P.I.E.S) analyses a live transcript of the conversation to improve customer experience by allowing service desk operators to concentrate on the human interaction rather than data collection. The system processes all information from the conversation in real time and enters it on the system while the call operator concentrates on the customer’s welfare. All information extracted can be verified by the call handler ensuring vital information is never missed or incorrectly identified. The system uses a BERT model trained on the Stanford Question Answering Dataset (SQuAD). One of the main advantages of the implemented system is the ability to produce more training data with every call the company handles. With thousands of calls per week, it is possible to generate a sufficiently large labelled dataset of transcripts specific to the company’s requirements. This allows for routine updating of the Question and Answer model to ensure it is performing to a high standard.
Multiplayer Rhythm-Based Dungeon Crawler
Games Project - Awarded Best 3rd Year Group Project
For a 3rd Year group project, we created a game called Rave Cave. It is a multiplayer rhythm-based dungeon crawler where large amounts of players rock out simultaneously in time to the music. Players must cooperate with their team to solve puzzles and complete complex button sequences - all in time to the beat of the music. We created our own game engine in C++ by integrating our custom code with external libraries.
Gold Crest Award & Engineering Education Scheme
The Royal Navy
The Engineering Education Scheme is a scheme run by the Engineering Development Trust, which links teams of 4 sixth form students with companies to work on an engineering problem. Our company was The Royal Navy and, over a year, we developed a replacement for a leaking Mantlet Bag. Our solution has been taken on board by the Ministry of Defence, and they have assigned their lead contractor to put our solution to further testing.
Talk at University of Bristol, Bristol, UK
Talk at Insight into University - Engineering Development Trust, Bristol, UK