One of the clear issues facing the Access the Justice movement in the UK is the apparent complexity of legal language. For those without legal training, this can be a serious barrier to searching and navigating legal documents. Whilst the National Archives' Find Case Law service provides free public access to judgments, its traditional keyword search system limits users to exact lexical matches, missing semantically similar legal concepts expressed through varied terminology. A search for "parental responsibility" will not retrieve judgments discussing "legal guardianship" or "custodial obligations," despite their conceptual equivalence. This is a critical limitation for self-representing litigants or anyone without a legal background wishing to conduct legal research.

This presentation demonstrates a working prototype that addresses this gap through vector-based search. The system employs a pre-trained transformer model to encode judgments as vectors, capturing semantic meaning rather than keywords. When users enter natural language queries, the system retrieves relevant passages through vector similarity comparison, surfacing conceptually related legal reasoning regardless of wording. Moreover, through the addition of metadata topic tags, the system displays results in a way that is clear to non-legal users by giving a high level overview of the themes explored in each case, matching these to the given search query.

You will receive the link to join the talk remotely or the details of the location approximately one week before the seminar.

Speaker:

Caitlin Wilson is a PhD candidate in Digital Humanities at King’s College London, holding a UKRI scholarship from the London Arts & Humanities Partnership (LAHP). She is working on a Collaborative Doctoral Award with The National Archives, working with their Find Case Law service to develop a semantic search function which will aim to reduce barriers to access by bridging the gap between legal and everyday English. She holds an undergraduate degree in Linguistics from the University of Edinburgh and a MSc in Digital Scholarship from the University of Oxford. She has also trained as a Digital Archivist at the UK Parliamentary Archives.