nlp - algorithm to get topic / focus of sentence out of words in sentence -
There are no well known or successful algorithms for obtaining topics and / or sentences from the focus of a sentence (question) The word in question?
If not, how do I get to know the topic / focus of the question? It appears that the subject / focus of questions is usually a noun or a noun phrase.
So the first thing I have done is to determine the nouns through the speech of the part of the question. But then how will I know that I should only have a noun, or noun (special) and before that an adjective, or earlier noun or action statement, or what the noun (names) and verb should be?
For example: Get 'Quick Brown Fox', 'Junk' and 'Lazy Dog' in 'Instant Brown Fox Jump on Sloth Dog'.
"What is 'Japan's population', 'Population' and 'Japan'?
In 'What is milk' find 'color' and 'milk'
In the 'What's the height of Mount Everest' Mill 'Mount. Eastst' and 'Height'.
While writing them, I think the easiest way is to stop the words. This can be considered as a parsing problem and I personally Tool swept The Stanford Parser has a link to the demo For example, did the quick brown fox jump on the lazy dog you write from the output To remove the nouns, an extractor (adjective and functionality, if necessary) and this In addition, (J.J. Brown) (N.N. Fox (VP jump) (PRT (RP)) (NP (DT) (J.J.L.L.) (NN dog) If you look closely at the pars tree, then you expect the production He is both NP (noun phrase) - quick brown fox and lazy dogs. I hope it helps!
done / vbd / dt quick / jj brown / jj fox / nn jumping / vb over / rp / dt lazy / jj dog / nn
(Root SINV (VBD done) (NP (D. D.) (JJ Quick) gets the topic from sentences.
Comments
Post a Comment