2005
DOI: 10.1080/02572117.2005.10587246
|View full text |Cite
|
Sign up to set email alerts
|

The effectiveness of morphological rules for an isiZulu spelling checker

Abstract: This paper shows how morphological analysis contributes to solving the challenges posed by the development of a spelling checker for an agglutinative language like isiZulu. It demonstrates how the incremental implementation of affix removal rules can be used to derive word forms and enhance the lexical and error recall of the system. In the case of the spelling checker the strategies used are mainly based on the use of regular expressions, and more specifically on a process of stemming.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
11
0

Year Published

2017
2017
2020
2020

Publication Types

Select...
3
3
2

Relationship

0
8

Authors

Journals

citations
Cited by 29 publications
(11 citation statements)
references
References 6 publications
0
11
0
Order By: Relevance
“…This being the case, the 'add word' feature of the spellchecker is an exciting avenue for further investigation into new words that are being added to the lexicon, and thus may soon provide a wealth of evidence on the intellectualisation of the language. The 'add word' feature is unique to this isiZulu spellchecker compared to earlier attempts (Bosch & Eiselen 2005;, which is providing a wealth of information for spellchecker development as well as linguistic analyses. A controlled test setting with university students might assist in obtaining such results.…”
Section: Discussionmentioning
confidence: 96%
See 1 more Smart Citation
“…This being the case, the 'add word' feature of the spellchecker is an exciting avenue for further investigation into new words that are being added to the lexicon, and thus may soon provide a wealth of evidence on the intellectualisation of the language. The 'add word' feature is unique to this isiZulu spellchecker compared to earlier attempts (Bosch & Eiselen 2005;, which is providing a wealth of information for spellchecker development as well as linguistic analyses. A controlled test setting with university students might assist in obtaining such results.…”
Section: Discussionmentioning
confidence: 96%
“…While efforts have been documented to develop spellcheckers (please see especially Bosch & Eiselen 2005), these tools are not available. The plugins for OpenOffice, Firefox, and Thunderbird -developed by translate.org.za in 2008 -are freely available, but they have not been updated so they no longer work with the latest versions (since OpenOffice v4.x).…”
Section: Human Language Technologiesmentioning
confidence: 99%
“…which is a subset of the conjugation (ba-for 3rd person plural) and CARP (-el-) and no details of its implementation is provided [3]. A related work on morphological rules focuses on nouns [23].…”
Section: Related Work On Isizulu Verbsmentioning
confidence: 99%
“…It covers only a fraction of the complexities of the isiZulu verb; for instance, it addresses only one extension to the exclusion of the causative, applicative, and the reciprocal extensions. Other attempts to formal approaches to isiZulu morphology focus predominately on nouns rather than verbs [23,24], or describe only a few sample regular expressions that cover a very small fraction of the verb [3]. The morphology of the verb is widely regarded as the most interesting theoretically.…”
Section: Introductionmentioning
confidence: 99%
“…Corpora in Africa are, obviously, not only about variational studies of English. They have supported the development, enhancement, or evaluation of a broad range of natural language processing tools in Africa: a speech synthesizer in Ibibio (Ekpenyong et al 2014), part of speech tagging in Bantu languages (de Pauw et al 2012), a statistical English-Luo machine translation system (de , spell checkers, as well as extended applications such as language identification in Gikuyu, isiZulu, Tiv, Swahili, Hausa and Igbo (Akosu & Selmat 2013;Chege et al 2010;Bosch & Eiselen 2005). The use of corpora for the development of predictive texting and translation memory solutions in other environments has also been described (Antia 2009).…”
Section: Introductionmentioning
confidence: 99%