Creating a common multilingual lexicon to enhance communication in digital evidence interactions

• In Event: Clarus: Culture, Bias, Language and Communicaiton in Digital Evidence Investigations

Abstract

This paper focuses on the third work package of the CLARUS project that aimed to create a common lexicon of written and spoken interactions of police and forensic science professionals in digital forensic examinations in five European countries — Finland, Czechia, Portugal, Greece and the UK. Glossaries have been created by the forensic science community using expert knowledge for some time, an example of which is the ISO/IEC guide 27042:2015 (ISO 2015) or the US-based Organization of Scientific Area Committees (OSAC) in 2018 (see Le Gall et al., 2022); however, the general lack of agreed formal language definitions with multiple language equivalents, remains a significant issue in a European context. In this project, we took an innovative multi-method and data-driven approach to lexicon construction that involved natural language processing, corpus linguistics techniques (the use of computer-aided language analysis to identify recurrent patterns), and qualitative analysis of multiple data sources. These multiple data sources involved written documents, including legislation, manuals, reports, and codes of practice that are distributed among police officers and forensic scientists in five countries, practitioner diaries, interviews and focus groups that were collected in the first two work packages of this project. This paper will provide an overview of our methodology to the lexicon, our challenges of working in partnership with practitioners, including police, forensic scientists, and academics across disciplines like criminology and psychology, as well as issues with translation and metadata collection. We will discuss the potential uses of the lexicon and its potential impact on digital forensics.

Author

• Duygu Candarli, University of Southampton