A competitive edge in the job market, thanks to language technology
A year and a half ago, ChatGPT hit the market and language technology now seems to be everywhere. Apparently very suddenly – but that’s not actually true. Research into language and AI has been going on for decades, and we have all been using language technology tools for longer than we realise. VAIA spoke to Professor Véronique Hoste and Doctor Cynthia Van Hee about the history of language technology, the importance of this discipline for employees and employers, and approachable ways to upskill.
“ It’s in everyone’s best interest to learn about language technology! ”
The history of language technology and AI
Véronique Hoste is professor of language technology at Ghent University, where she heads research group LT3 (Language and Translation Technology Team). ‘Language technology is trying to understand and generate language automatically,’ she begins. ‘Imagine a computer programme that “understands” what you mean, and that can also produce text or speech.’ ChatGPT may be a prime example, but language technology has been around much longer than this particular tool. Véronique explains that language and artificial intelligence have been strongly linked since the 1950s, and the history of AI has always been propelled by advances in language technology research.
‘Research in this field is indeed not new,’ Cynthia Van Hee continues. She is a postdoctoral researcher and lecturer in language technology at Ghent University. ‘But in recent years we see a huge boom. This is due to more funding for the research domain, more technological possibilities through more powerful computers, and also the large amounts of data available with the rise of social media. That combination has enabled the development of powerful tools like ChatGPT that are accessible to the general public. Thus, language technology is more visible than ever today.’
Véronique adds: ‘But make no mistake, for years we have all been using a lot of systems that are made possible by language technology. Just think of search engines like Google, which have been around since the late 1990s.’
Language technology in the workplace
A lot of companies have jumped on the language and AI train, notes Véronique. But sufficient background knowledge is often lacking: ‘Many people start working with such tools without knowing what is going on under the hood. That gets in the way of critical thinking. Whereas with a little knowledge, you can use the technology much more efficiently!’
This is why both experts stress the importance of educating yourself in language technology. ‘It really gives you an edge in the job market,’ Cynthia says. ‘Whatever sector you currently work in or want to end up in. Language technology offers different opportunities per domain or company: this can range from deploying chatbots to optimising business processes, for instance based on consumer behaviour. To properly assess those opportunities, employees and employers need to know how the technology works. This is also crucial for responsible use. It is important that users of, say, machine translation tools know that the output may contain biases or stereotypes. But equally important that they realise what happens to their data.’
‘Moreover, language technology systems are sometimes misused,’ adds Véronique. ‘A major pitfall for today’s generative AI systems is disinformation. AI-generated content is becoming increasingly difficult to recognise because the language seems so natural. Creating awareness around that is super important.’ And this should preferably start early, while still in school – but for now, language technology is only taught in a few subjects in secondary school and higher education. But anyone who is already working can retrain themselves in this topic. And rest assured: there is no need to take a long course. Véronique agrees: ‘To learn how, for example, chatbots based on large language models such as ChatGPT work, a one-day course or an understandable book can be a great starting point!’
A book for a wide audience
And that's exactly the kind of book that Véronique and Cynthia released. They are editors and co-authors of the recently published Taaltechnologie ontrafeld (Language Technology Unravelled). This first Dutch-language handbook on language technology zooms in on basic concepts from the discipline, specific applications and tools, and ethical and legal issues surrounding language technology.
Cynthia articulates the mission of this book: ‘We want to give people the right foundation as well as the self-confidence they need to get started and explore the topic further on their own. Think of this as a reference book for non-researchers. Although the authors are all scientific experts, the text is not academic but very accessible. Admittedly without compromising on depth.’
Finally, Véronique explains which type of reader Taaltechnologie ontrafeld is meant for: ‘Anyone starting from scratch will be able to understand the entire book. Although, of course, we sometimes examine some tougher, technical concepts. People who are already more advanced in language technology will definitely learn something new, for example about the history of language and AI, or about specific applications.’
‘As a reader, you can also freely navigate through the book,’ Cynthia adds. ‘If you already have some background knowledge, feel free to skip the first section and dive straight into how specific applications such as translation systems work. By the way, note that we highlight a very wide range of applications and methods, including hands-on demos. There is something for everyone!’
Dive into the book
Taaltechnologie ontrafeld is available from the publisher Pelckmans, Standaard Boekhandel, bol.com and FNAC. Eight co-authors contributed to the book: Véronique Hoste, Cynthia Van Hee, Orphée De Clercq, Els Lefever, Lieve Macken, Arda Tezcan, A. Seza Doğruöz and Joke Daems. They are all associated with the research group LT3 (Language and Translation Technology Team), Ghent University.
Take a course
Visit the VAIA course calendar to keep up to date on language technology courses. Some examples include: a micro-credential on natural language processing, or a postgraduate certificate on technology for language professionals. Prefer to learn at your own pace? Then be sure to check out Professor Ayla Rigouts Terryn’s video series on language and AI.
Veronique Hoste
Veronique Hoste is Senior Full Professor of Computational Linguistics at the Faculty of Arts and Philisophy at Ghent University. She is department head of the Department of Translation, Interpreting and Communication and director of the LT3 language and translation team at the same department. She holds a PhD in computational linguistics from the University of Antwerp (Belgium) on "Optimization issues in machine learning of coreference resolution" (2005).
Natural language processing in the last decades has drastically shifted from a discipline in which linguistic rules were handcrafted by experts to a vibrant discipline heavily dominated by machine learning and more recently, end-to-end neural approaches. Véronique has a strong track record in combining linguistic expertise with machine learning of natural language, seeking to automatically model natural language up to the level of semantics and discourse in a wide range of applications, such as cyberbully detection or sentiment and emotion analysis.
Cynthia Van Hee
Cynthia Van Hee is a lecturer and postdoctoral researcher at Ghent University, affiliated with the LT3 Language and Translation Technology Team. Her work as a postdoc centers on computational linguistics and machine learning, with a focus on sentiment analysis, irony detection, and linguistic preprocessing. She teaches several courses related to language technology, including Data Processing with Python and Natural Language Processing, as part of the Applied Linguistics programme and the Computer-Assisted Language Technology postgraduate.
Lisa Hilte
Lisa Hilte is doctor of linguistics. For years, she worked as a linguistic researcher and lecturer at the University of Antwerp. Today, she works for VAIA as an AI expert and liaison for UAntwerp. In addition, she is a freelance writer with a focus on science journalism, and a freelance teacher in AI and language technology. She has published two sets of course materials on language technology, for pupils and teachers in the third grade of secondary education (Plantyn publisher).
