The Multilingual Semantic Web as Virtual Knowledge Commons: The Case of the Under-Resourced South African Languages

Abstract

The participation of the under-resourced South African languages in the Multilingual Semantic Web as Virtual Knowledge Commons is imperative in terms of sharing in and contributing to the knowledge commons, in sustaining multilingualism and the technological development of these languages and in preserving cultural diversity and indigenous knowledge systems. This chapter takes a closer look at the challenges that the under-resourced languages of South Africa face in this regard and addresses two of these challenges. It is shown how three different types of high-quality language data, viz. multilingual terminology in English, Afrikaans, Tswana and Zulu; indigenous knowledge on astronomy nomenclature in Tswana; and a parallel corpus of English, Afrikaans, Tswana and Zulu could be exposed as Linked Data in a principled way. The conclusion contains various possibilities for future work.