Defining user intentions in a multi-touch multi-user application

I am developing a surface application (although the platform is not really relevant) that should track usage statistics and create data that can be analyzed to answer a few usage related questions.

In particular, I have a screen in my application that displays several pieces of content at once (ScatterView for superficial people). On this screen, several people can simultaneously interact with several pieces of content.

I need to answer two specific questions:

  • What is the most / least popular piece of content in the last 24 hours.
  • How long each piece of content was viewed for 24 hours.

I am struggling to find a conceptual approach to this problem. I make some assumptions that, I think, simplify the multi-user problem.

  • Any contact oriented between 0-180 degrees is a user 1
  • Any contact oriented between 181-360 is a user 2
  • After a period of inactivity (no contacts), the current session will end. This gives me the ability to distinguish between user sessions and track usage time.

My problem is how can I (even somewhat reliably) determine user intentions regarding a piece of content?

  • How do I know that they are “browsing” part of the content?
  • How can I determine if something is “popular”?

Any ideas on how to approach this would be appreciated (no matter how they look on the wall)

UPDATE:

. "" .

, TokenManagement. ApplicationToken, UserToken AssetToken.

ApplicationToken .

UserToken (0-180 181-360). , , . ( 1 ), . , , . ( ?)

AssetToken , , ( , , AssetToken). , , , .

,

  • ( UserToken)
  • (Max of Distinct AssetTokens )
  • (Min of Distinct AssetTokens )

?

+3
3

, - , . , "", . , , .

, - (, , -).

, , .

- .

+1

( , , , . , , , ). . , , ? ? , , , .

, . , - , .

+1

, Surface, 10- amazon mechnical turk.

.

, - , , .

, .

+1

Source: https://habr.com/ru/post/1723715/


All Articles