Question

I'm planning a Social Network with an Artificial Intelligence . that means that the SN will take the conversations (in english) .. and analyse them in order to extract the general opinion about a subject. This helps to collect information and build statistics, which will be sent then to the appropriate user. My question is : how to organize words and grammatical rules in a database, in order to help the social network extracting a general opinion from a conversation (agree, disagree..etc)! thank you.

Was it helpful?

Solution

its very challengning task to extract text that reflects real opnions of people because machine learning is still not advanced in this area, take a look at the following example: your program finds this sentence on some SN website: "I made her duck"

what are possible options that your program will infer:
1-i cooked a duck for her
2-i "magically" transformed her into a duck"
3-i took her duck and created it.
and many other

so its really challenge. I suggest to take a look at NLP(natural language processing) which covers this area you're interested in

OTHER TIPS

Could you plug Wolfram Alpha (http://www.wolframalpha.com/about.html) into your site? That would save probably a LOT of work.

Good luck with that :-) 15 years ago I did a 20 week course at university which barely scratched the surface of the question you ask. Start by googling 'natural language interface'...

Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow
scroll top