consider looking into how IBM developed Watson. There is information online. Dan
On Mar 4, 2016, at 2:38 PM, Nishant Kelkar <[email protected]<mailto:[email protected]>> wrote: This sounds like a mammoth task, you want concept recognition (e.g. Running is a type of <physical activity>). You'd probably need more than just a thesaurus as your auxiliary data source. Try searching DBPedia for how to get a large set of name --> type mappings, it's available for free. Just curious, since this is a hard problem that you're trying to solve: do you have any particular type of questions in mind that you want answered, or just any random questions need to be answered as well? Best , Nishant On Friday, March 4, 2016, Vyacheslav Gonakhchyan <[email protected]<mailto:[email protected]>> wrote: Hi, I want to write a program using apache opennlp library. Program should take input and thesaurus(synonyms, antonyms, etc.) and produce a program that answers questions. E.g. Input: input.txt containing Q&A, thesaurus.txt. Example input.txt: Q: Do you like apples? A: Yes. Q: Do you like running? A: Yes. Example program run: Q: Do you like fruits? A: Yes. Q: What activities do you prefer? A: Running. Is this Q&A program possible with opennlp? Totally new to this powerful software. Please point me in the right direction.
