I am working on Tamil grammar rule based sentiment for Tweets (Tamil Movies). How many grammar rules are there in Tamil and how to use those grammar rules for Tweets for categorization.
Hi Nadana Ravishankar,
There are prepositions, negation, questions, adverbs, and pronouns as important rules in Tamil. As I believe that you just follow the structure of the language is enough since you are going to type limited number of letters.
Hi,
I don't think anyone can say how many grammar rules there are in any given language. Theoretically, whether you are a generativist grammarian or just an NLP practitioner, the assumption is that there is a finite number or grammatical rules (though we don't know how many) capable of analyzing or generating an infinite set of well-formed sentences.
In practice, though, it all depends on a lot of variables. I think the most important one is the application. If you're trying to achieve fine-grained syntactic representations (for example, if you're building a Tamil Treebank, or if you're working on deep parsing for Machine Translation), then you'll probably need rather complex syntactic formalisms (by descending order of complexity: HPSG, TAG, LFG, PCFG), and you'll have to manage quite a few rules, and probably end up writing a PhD in the process. On the other hand, if you're "just" trying to spot specific words or patterns, regular expressions or finite state transducers will generally do the trick. Now, if Tamil allows for complex morphological operations such as can be found in semitic languages (i.e. 3-consonantic roots, eg. k-t-b --> kitab, kataba, mektub...) or if word boundaries are not that clear (as in Arabic, or worse Thai), then you'll probably have to use weighted finite-state transducers, just for spotting words. So even spotting patterns made of specific words might not be that easy.
Could you elaborate a little on the kind of application you're working on?
I am also doing this research, also waiting for more answers
தமிழ் மொழியில் 100க்கும் மேற்பட்ட இலக்கண விதிகள் இருப்பதாக கூறப்படுகிறது. ட்வீட் வகைப்படுத்தலுக்கு அவற்றைப் பயன்படுத்த, ட்வீட்டின் சூழலில் ஒவ்வொரு விதியையும் எவ்வாறு பயன்படுத்துவது என்பதை நீங்கள் புரிந்து கொள்ள வேண்டும். தமிழ் இலக்கணத்தின் அடிப்படைகள் மற்றும் ட்வீட்களுக்கு அது எவ்வாறு பொருந்தும் என்பதைப் புரிந்துகொள்ள உதவும் பயிற்சிகளை ஆன்லைனில் நீங்கள் காணலாம்.
Tamil grammar and how to use it for tweet categorization: Tamilcube, Etsyakalam, and Tamil Language and Culture
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude