Dr daniel kershaw

Name: Dr Daniel Kershaw (2012 cohort)

CDT: HighWire – Lancaster University

Nationality: British

Status: Alumni

Research area: Online Social Networks and Language Change

Current role: Senior Data Scientist at Elsevier, London


Daniel joined the HighWire CDT after completing a BSc in Computer Science at Lancaster University. He chose the HighWire programme as he wanted to further his studies and to pursue his interest in research-based computing roles in industry.

His PhD research focused on changes in online social networks. This involved taking large samples of data from online social network sources (from Twitter and Reddit) and developing large scale computational models which extracted social networks and modelled language change (the emergence of new words) across the networks between users.

Daniel produced a number of publications from his research and presented talks around the UK and internationally, including in South Korea. In addition, his research was featured in the New Scientist Magazine in January 2016 – see newscientist.com/article/dn28787-tweets-and-reddit-posts-give-snapshot-of-our-changing-language/.

Daniel found that the work life balance of a PhD was sometimes challenging, and acknowledged he could not continue working as he had done during his undergraduate years. However, he developed an ability to recognise when something was completed, and knowing when it was time to reflect on work.

His advice to someone embarking on a PhD is to be prepared for the unexpected! Through being self-motived during the PhD and being able to structure modelling language problems independently, Daniel was able to adapt modelling language problems for business use.

Daniel has now relocated with his partner to London and has a successful career as a Senior Data Scientist at Elsevier. He believes that his research shows that language change can be successfully tracked using the internet and means that we should, in the longer term, be more able to understand emerging trends in new word usage.

%d bloggers like this: