SADiLaR Language Resource Repository
Not a member yet
536 research outputs found
Sort by
Afrikaans Wordnet 1.0
No full textThe Afrikaans WordNet is a lexical reference source displaying some similarities with dictionaries and thesauri. Together with ALEXANDER, our other Afrikaans database, it should be of great use as a reference source and application. It can easily be linked to other wordnets for translational purposes. This also provides a rich source of research opportunities, because different languages are compared to each other conceptually. The Afrikaans WordNet can also be used as source in the development of technologies to disambiguate the meanings of words automatically, for document categorisation and summarisation, as well as machine translation. Entries consist of sets of (one or more) synonyms (so-called "synsets") that refer to the same concept, such as "tafel" (Princeton WordNet: "table"). Synsets are linked to each other through different semantic relations, such as the hyponymic relation which implies that "tafel" is a kind of "meubelstuk, meubel" (Princeton WordNet: "furniture, piece of furniture, article of furniture"). By following the relations, one is able to move from "tafel" to "meubelstuk, meubel" in the database. All the synsets in the wordnet are connected via different semantic relations, resulting in one big network of concepts
Hyphenator 1.0.
No full textRule-based hyphenator which can be implemented in any NLP system. It takes as input a string, and produces as output an analysed string, without any tags. For example, the string "hondehokdak" ('dog house roof') will be analysed as "hon-de-hok-dak", where the hyphen indicates syllable boundaries on an orthographic level. Rules implemented in Perl
African Speech Technology isiXhosa Text Corpus
No full textMonolingual text corpus developed during the African Speech Technology project
Sepedi Tokeniser
No full textPre-processing for Sesotho sa Leboa morphology as a disjunctively written language. (morphemes are already separated) as pre-cursor for morphological decomposition
Combination Tagger
No full textThe combination tagger framework uses MBT, SVM, MXPOST and TnT. Each tagger receives a weight by which it can vote for a tag
Intonation model for Bantu tone languages
No full textTheoretical linguistic intonation model for Setswana-Sesotho and isiZulu, formulated as ste-wise rules in prose
Fannie Sebolela Oral Corpus
No full textTape recordings and transcriptions of 13 mother tongue speakers. Transcribed orthographically
