Syllable-Based Text Compression: a Language Case Study

Abstract

Compression of texts has been widely studied by various researchers and in the process, several algorithms have been proposed. However, compression of texts using the syllabic structure of words in syllable-based languages has emerged as another dimension to the compression of texts. An algorithm for syllable extraction from words should be designed based on the structure of a language due to the ineffectiveness of the presently existing "universal" algorithms. Several syllable-based methods of compression proposed by different authors are reviewed in this work, including the methodologies used in achieving text compression. Finally, an algorithm for syllable extraction from words in the Yoruba language is presented and compared with four universal algorithms, recording the best result (100 % accuracy) among the five; the significance of this is that a dictionary of common syllables does not need to be created to achieve syllable-based text compression on the Yoruba Language.