Hello everyone! It is finally spring here in Boston. We have warm weather, beautiful Magnolias in the bloom, and warm sunny days. Magnolias in bloom.
It is hardly good to be the only smart person in the room; today I started a wiki to collect all the good work done by several academics and industrial scientists, engineers in the field of spell checking.
In this article I will present a conceptual framework, a proposal, for writing a multi-pass solthiruthi spell-checker. This effort is variation on theme of other spell checkers for Tamil created by projects myspell, hunspell, navi/vani, and languagetool respectively.
N.B.: I invite interested individuals to contribute to this document, and collaborate with me, under usual open-source credit/attribution terms.
Given a Tamil text, urai, we want to find out all spelling mistakes in the text
For words which are in error, we want to further provide alternatives
For Tamil language we have to apply santhi rules, puranchi vithigal, from grammar.
We also would like to provide context sensitive help like A.I.
I propose a multi-pass spell checker comprised of N-levels. Each level upto the penunltimate (Level N-1) will continue to refine the words which are found in error; we call this analysis levels. The final level will form the suggestions for the words in error; we call this synthesis level. Such an algorithm can be built from 2-Level and upwards in the complexity.
We use a dictionary of valid Tamil words, to validate the Incoming text is split into the form of words, and each word goes through the N-steps.
Typical levels for analysis include, but not limited to,
Dictionary based white-list analysis for incorrectly spelled words
Morphological analysis via stemmer – this is a requirement of agglutinative languages
Clean up proper nouns words, peyarechcham, using names of places, people, things
Bayesian classification of errors using bigram, trigram, N-gram letter probabilities
Context sensitive word suggestion by Bayesian classification using N-gram word probabilities
Keyboard models, Bamini, Tamil99, Anjal, layouts and common misspellings in these
Edit-distance based analysis for mistyped words
Detect transliterated English words in Tamil and ignore
Synthesis step is usually made from the following,
For Tamil we need to combine Santhi, and Punarchi rule based pruning of suggestions
Pizhai : 2-Level spell checker
Pizhai spell checker algorithm uses the 2 analysis steps of whitelisted dictionary and the bigram/trigram letter analysis to flag erroneous words. Finally we suggest word alternates to the user during synthesis.
Solthiruthi : 3-Level spell checker
Solthiruthi spell checker algorithm uses the 3 analysis steps of whitelisted dictionary and the bigram/trigram letter analysis to flag erroneous words, and includes a stemmer. Finally we suggest word alternates to the user during synthesis.
All of these ideas can be implemented in Python using open-tamil project, today. We welcome enthusiastic contributors – linguists, language enthusiasts and developers.
இன்று நள்ளிரவு எழுதிய பைத்தான் நிரலிநால் (இந்த கிட்ஹப் கமிட்டை காணவும்) எழில் மொழியை இணையம் வழி கற்க பள்ளிக்கூடம் ஆக அமைய வாய்பு உண்டு. இதனுடைய அமைப்பு பல விஷயங்கள் கொண்டது. கீழே காண்க.
Code in this directory provides the following
* Writing Code, Editing, and Evaluating *
2. Code browser lets user to look at sample Ezhil programs from the ezhil-lang source/testsuite, in the single page app editor
3. Users can run the code on this page, and see the output in the same page.
4. Correctly executed code with should output in light yellow; clicking on the output will hide it, as you work on second problem.
5. Errors in code or server execution cause your program output to be highlighted in red.
6. Source code is persisted between sessions in terms of cookies