Profile
Biography
Sam Relins is a Research Fellow at the ESRC Vulnerability & Policing Futures Research Centre, specialising in data science with a focus on machine learning and natural language processing. A graduate of Mathematics & Accounting & Finance from Sheffield University, Sam spent several years working in the finance industry as a stockbroker.
Research
Overview
Sam’s research aims to enhance the understanding of police interactions with individuals facing mental health challenges, with a focus on analysing free-text police incident logs. Traditional data collection and statistical methods often miss the depth of information contained in these logs, but Sam aims to leverage recent advances in Large Language Models (LLMs) to uncover these hidden insights. His primary focus is using Large Language Models (LLMs) to generate synthetic police incident logs, replicating sensitive data while maintaining privacy. Using these generated logs, he is exploring the potential to fine-tune LLMs for accurately classifying genuine police incident reports, or to leverage instruction-tuned LLMs to classify reports in zero or few-shot settings. This work is not only theoretical but aims to directly impact the way police handle data, harnessing the power of LLMs to interpret textual information.
