1
$\begingroup$

As a part of my project, I have to ask survey questions to a set of people. I would like instead of a static survey where everybody is asked the same questions, but for the questions to be tailored to the user in one of the following ways : 1. Based on a set of questions that I ask at the beginning (creating some user profile) 2. May be also based on the answers he/she provides to the question.

I want to do this so I can explore interesting aspects for each user rather than the same questions for all.

I know this is a very abstract idea,but I wanted to ask the community if this is possible using Neural networks? If so, can somebody guide me in the right direction? I did not find much material on this except for the 20 questions game.

Thank you.

EDIT :

For example, if I ask a person, to who they are comfortable sharing their mobile phone sensor data, and they say they arnt comfortable sharing it with corporations, id like to ask them if they would share it if I gave them money, or if they would share it if i told them it was used somewhere,for example creating an health app that might be useful for them (this is just an example).

| cite | improve this question | |
$\endgroup$
  • 1
    $\begingroup$ What do you mean by "interesting aspects"? Do you mean things like not asking questions about a person's current employer if they indicate that they're not currently employed? (I.e. skip inapplicable questions, as you might indicate on a written survey.) Or something more dynamic. My fear would be that something more dynamic is itself based on a model you have (perhaps only in your head) and that you might end up with results that are so dissimilar that you can't really draw conclusions. $\endgroup$ – Wayne Mar 17 '16 at 17:11
  • $\begingroup$ i edited my question $\endgroup$ – LoveMeow Mar 18 '16 at 10:01
  • 1
    $\begingroup$ Neural nets take tons of data to train accurately. Probably not the best way to go about it. $\endgroup$ – Frobot Mar 19 '16 at 10:50
  • $\begingroup$ could you suggest another way to go about it? $\endgroup$ – LoveMeow Mar 21 '16 at 9:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.