Singapore Mandarin Usage Dictionary: A Computational Lexical Infrastructure for Localised Language

This project aims to develop the most extensive open-access, bilingual online dictionary to date focusing specifically on localised vocabulary in Singapore Mandarin, a computational lexical infrastructure that systematically identifies and analyses region-specific words and usage patterns.

Integrating linguistic theory, large-scale corpus analysis, and advanced natural language processing (NLP) techniques, the project addresses significant gaps in existing resources and offers insights into how language reflects and shapes Singapore’s unique sociocultural and multilingual environment. It provides an empirical basis for re-examining theories of lexical variation, language contact, and the development of the World Chinese framework (modelled on World Englishes and highlighting the global diversity and localisation of Chinese varieties, while also delivering a practical tool to support education, translation, media, and NLP development in a linguistically diverse context.

The project is structured around three interrelated objectives:

• To develop NLP methods to automatically detect localized lexical items in Singapore Mandarin, offering a scalable, data-driven alternative to traditional dictionary compilation;

• To analyse the lexical differences that distinguish Singapore Mandarin from other Mandarin varieties, and to use these patterns to advance theories of contact-induced change in multilingual settings and contribute to the development of the World Chineses framework; and

• To design and launch a user-focused bilingual dictionary platform that supports education, translation, media discourse, and NLP applications, while promoting culturally grounded and locally relevant language use in Singapore Mandarin.