Academics present research on Language technologies, AI and visualisation at the 2020 Wales Academic Symposium on Language Technologies

Academics from the School of Computer Science and Electronic Engineering present two research projects at the that took place on the 4th November 2020. The focus on the symposium was on Language Technologies, including Speech Technology and Translation Technology; Natural Language Processing; and Artificial Intelligence and Language.

Wales Academic Symposium on Language Technologies 2020Wales Academic Symposium on Language Technologies 2020

 

 

 

 

 

 

 

 

Professor Jonathan C. Roberts and Dr Peter Butcher presented their talk titled 鈥淒eveloping ColloCaid 鈥� lessons learnt in developing a text-editing tool, and visualising words, to help writers with collocations鈥�. Professor Roberts said 鈥渋t was good presenting the ColloCaid project and tool at this event. The project is funded by the Arts and Humanities Research Council and is a collaboration with researchers from 麻豆网, and and is led by Surrey University. Our presentation was divided into four parts. We first gave background information about the project and the data, second presented how we had designed and developed ColloCaid, third discussed some lessons learnt, and finally we gave a demonstration of the latest version鈥�.

Dr Butcher said 鈥淚t was good presenting a live demonstration of the tool at this symposium and it was great meeting other researchers in the area of language technologies. The ColloCaid tool has a huge underpinning corpus of collocations, with over 30,000 words and example sentences. Users can write their own texts in one window, where the text analysis system analyses the words, and drawing from our large corpus of collocations, the tool suggests appropriate collocated words. In a second linked-window, we visualise these collocations to allow people to explore alternative words in more depth

Peter Butcher and Jonathan C. Roberts presenting the ColloCaid project, showing the text editor (left) and one of the word visualisations (right)Peter Butcher and Jonathan C. Roberts presenting the ColloCaid project, showing the text editor (left) and one of the word visualisations (right)

 

 

 

 

 

 

Dr William Teahan and Leena Farhat (a Masters student at 麻豆网) presented 鈥淎dapting the Tawa toolkit for modelling and processing Welsh text鈥�. Dr Teahan said 鈥淭awa is a compression-based toolkit for processing and modelling information. It follows the idea that applications can be considered as problems in searching for the best encoding. We have been applying the idea to natural language processing, and recently to the Welsh language. We were delighted to present the research at the symposium鈥�.

Dr William Teahan presenting the Tawa toolkitDr William Teahan presenting the Tawa toolkit

 

 

 

 

 

 

Professor Roberts finished by saying: 鈥淭here are many challenges and opportunities for research in text mining, corpus linguistics, natural language processing, AI and big-data visualisation. As academics and researchers from the School of Computer Science and Electronic Engineering we were pleased to be part of this symposium and present our research. Personally, I also enjoyed the round-table discussion, which took place at the end of the day. Many researchers wanted to follow up on both research projects, and we discussed generally of how to collaborate with research in the area of language technologies and natural language processing. In our School, we are always willing to collaborate across academia and industry, and look forward to developing new projects in AI, NLP and visualisation鈥�.

Publication date: 11 November 2020