UMQAGELI

( AUTOMATIC IDENTIFICATION OF BANTU LANGUAGES )

Save the text
Umqageli means 'guesser' in zulu. It is a program that takes the text (at least 15 words) you submit and guess in which Bantu language it is. The longer the submission is, the more accurate the answer of Umqageli is. But I think that this program will become really efficient when the number of supported Bantu languages is about a hundred. It will prevent diagnosis errors that occur when a non-supported Bantu language is submitted to Umqageli. This program, written in PHP, is an implementation (adapted to Bantu languages) of the algorithm of Cavnar, W. B. & J. M. Trenkle, 'N-Gram-Based Text Categorization', 1994.

--- Jacky Maniacky

Please enter a text (minimum of 15 words) in the field above

Linguistic tools