Antal van den Bosch @ LEL Research Seminar

Posted on February 13, 2018 by


On Tuesday 20th February our department welcomes Prof Antal van den Bosch to the LEL Research seminars. Antal, affiliated with the Radboud University of Nijmegen but currently director of the Meertens Institute (Amsterdam), describes his research interests as follows:

“In my research I develop machine learning and language technology. Most of my work involves the intersection of the two fields: computers that learn to understand and generate natural language. Specific interests include memory-based learning, machine translation, the relation between written and spoken language, text mining, the Dutch language, computational humanities, and cultural heritage.”

In his talk (see abstract below), he will discuss the computer’s answer to the question of linguistic form, structure and meaning.

As in the last three weeks, the talk will take place at Sam Alex_A101, from 16:15 – 17:30.

Afterwards, there will be a short reception and we will take our speaker out for dinner. We look forward to seeing you there!


The computer’s answer to the question of linguistic form, structure, and meaning

Prof. Dr. Antal van den Bosch

(Meertens Institute, Radboud University Nijmegen)

The field of computational linguistics seems to have little to offer to other branches of linguistics because of its engineering goals and fully orthogonal views on linguistic models and knowledge. However, its products do contain linguistically interesting information. Cutting open a phrase-based machine translation system, the type of MT system that Google Translate implemented until recently, reveals what could be seen as the computer’s answer to the question of linguistic form, structure, and meaning, to paraphrase Douglas Adams. In my talk I review ways in which machine translation phrase bases and other related statistical language models can be used for different, linguistically more interesting goals than for which they were produced: automatic paraphrasing, psycholinguistic and neurolinguistic experiments, and language variation.