Professor Andrew Kehoe
Professor of Corpus Linguistics / Director of Research
Andrew Kehoe is Professor of Corpus Linguistics and Director of Research in English. He studied at the University of Liverpool, gaining qualifications in both English and Computer Science. For over 20 years he has worked on and led a series of UKRI-funded projects in the field of Corpus Linguistics: the automated analysis of patterns and trends in large text collections to discover how language is used by different groups in society. Andrew leads the Research & Development Unit for English Studies (RDUES), an interdisciplinary team developing software tools which are used by hundreds of thousands of people worldwide to analyse textual data in research, teaching and commercial contexts.
Andrew has a particular interest in the analysis of online data, including news articles, blogs, social media and e-commerce. The RDUES team developed the WebCorp search engine with EPSRC and AHRC funding (2000-11) to allow users to extract examples of language use from such online data. WebCorp provides access to the Birmingham Blog Corpus, a 630-million-word collection of blog posts and reader comments.
Andrew also led the development of the eMargin collaborative text annotation tool, funded by Jisc from 2011-13. eMargin was originally designed for the close reading of literary texts but now has over 15,000 registered users across disciplines worldwide. More recently, Andrew was Co-Investigator on the ERC Horizon2020 RED-Alert project detecting online terrorist content, and Principal Investigator on the AHRC-funded TRAC:COVID project (2020-21). In TRAC: COVID, the team built a corpus of 84 million UK tweets and an open-access dashboard to allow researchers, policymakers and the public to visualise changes in the online conversation during the pandemic. The project produced case studies on vaccination misinformation and the clarity of official messaging, and its findings were cited in a House of Commons report on lessons learnt from the government's handling of the pandemic.
Andrew has co-edited two volumes on Corpus Linguistics, and has published a series of articles and chapters which have explored in depth the nature of web texts and the issues involved in extracting linguistic examples from them. Andrew has extensive experience in all aspects of research planning and management. In addition to his work on the WebCorp project, he was Research Associate on the APRIL neologism project (EPSRC, 1999-2000) and SHARES document similarity project (EPSRC,2001-4). After moving to BCU in July 2004, he was Researcher Co-investigator on the WebCorpLSE and Repulsion projects (EPSRC, 2006-8 and 2006-7 respectively) and co-author of an AHRC Knowledge Transfer Fellowship introducing WebCorpLSE to A-Level students (2009-11).
Andrew has worked as linguistic consultant to the Grey London advertising agency on a product launch campaign by Procter & Gamble, manufacturer of a new range of fragrances licensed under the Puma brand. The campaign was targeted at consumers aged 14-25 and aimed to raise awareness of the new product through a social media campaign. The agency’s idea was to allow consumers to write a message to a friend, which would then be translated into a video showing dance moves related to the content of the message. Andrew’s specific task was to determine which words were likely to occur most frequently in social media communication between young people. This research fed into the Dance Dictionary website, accompanied by television advertisements across Europe.
Andrew led the School of English return to REF2014 and REF2021.