Wmatrix corpus annotation and comparison tool: bring your own corpus!

Paul Rayson

SCC, Lancaster University

Wmatrix is a web based corpus annotation and retrieval tool which currently supports the analysis of small to medium sized English corpora and brings together techniques from corpus linguistics and natural language processing. In this two-hour hands-on workshop in the lab you will be guided through the main functions of Wmatrix which combine the keyness approach with corpus annotation to allow stylistic and conceptual analysis of texts loaded through the web interface. Wmatrix has been used for numerous studies in linguistics and social sciences (e.g. metaphor in end-of-life care, Shakespeare's comedies and tragedies, argumentation in reading groups, cultural change, formulaic language, language of psychopaths, political discourse analysis), medical education (e.g. problem-based learning transcripts), computer science (e.g. mining of early aspects and requirements in software engineering texts), management and financial studies (e.g. entrepreneurship, financial narratives). You will also hear about the USAS multilingual semantic tagger which will be incorporated into the next version of Wmatrix to extend the system to new languages. Please bring along your own corpus data in plain text or XML encoded form to test the system.

Week 24 2016/2017

Thursday 18th May 2017

Hannaford Lab