Leveraging non-expert semantic intuitions to support multilingual NLP
 

Olga Majewska, PhD student in Computational Linguistics, Faculty of Modern and Medieval Languages and Linguistics

The recent advances in Natural Language Processing have greatly increased the capacity of automatic systems to understand human language. However, they rely on the availability of large quantities of data, and still struggle with many language-related tasks which humans perform intuitively on an everyday basis. Making subtle meaning distinctions requires rich, fine-grained lexical-semantic and conceptual knowledge. Although automatic lexical acquisition systems promise to overcome the challenge of creating deep lexical resources manually from scratch, they depend on the availability of gold standard datasets for evaluation purposes, which is still very limited in most languages of the world. Verbs pose a particular challenge for NLP systems due to their complex linguistic properties. Acting as sentence pivots, they encode crucial information about the structural and semantic relationships between the elements of the clause. This is why accurate, nuanced analysis and representation of their meaning is especially important for NLP systems to get closer to human levels of language understanding.

Fast but reliable creation of semantic resources could boost and support multilingual NLP, eliminating the bottleneck of resource scarcity in the majority of the world's languages, and this project aims to facilitate this by developing methodology designed to speed up the resource creation process and allow its unlimited extension to diverse languages. It explores methods for obtaining verb classifications alternative to manual lexicographic work by leveraging semantic intuitions of non-expert native speakers. By examining humans' complex, intuitive word similarity judgments in different languages and encoding them in computer-readable form, the study explores how meaning relations are organised in the semantic space in the brain and provides insights to support further development of representation learning models and their ability to capture fine-grained semantic distinctions present in the mental lexicon.

 

 

 

Heritage language learners on the move: The transnational process of managing and learning Chinese in a Mandarin community school

Lini Xiao, PhD student, Faculty of Education

The growing segment of immigrant-origin children in the UK has directed scholarly and public attention to the complexity of heritage language (HL) education. Until recently much research and policy has conceptualised HL learners as a population permanently settled in a bounded society, with an aim to integrate into that society only. This assumption, however, is problematised by the changing nature of migration, which is increasingly recognised as a complex matrix of interactions and connections over time and space, rather than a linear movement across borders. This raises the question of how the language and educational needs of these increasingly mobile learners are being addressed.

Placed in the contexts of transnationalism and new Chinese migration to the UK, my project explores how transnational (im)mobility mediates contemporary HL learning in a Chinese community school in London. Following an ethnographic case study approach, the project involves intensive engagement with the community through questionnaires, participant observations, and interviews. The initial findings suggest that HL learners’ linguistic experience and learning trajectory are increasingly characterised by cross-border movements, multi-stranded transnational ties, and multi-layered desires for actual and virtual movements. It is in this shared condition of transnational mobility that knowledge, identity, and social relationships are (re)produced. Furthermore, with mobility being a key organising principle, Chinese HL education plays a key role in shaping learners’ repertoires for transnational participation and therefore constructing their future mobility.

This presentation will focus on a conceptual discussion of the entwined phenomena of global human movements and HL learning. I will draw on preliminary data to exemplify how HL learning, both at the community and individual levels, is organised in response to heightened transnational mobility. Some emerging implications for theory and policy will also be discussed.