I want to use machine learning to identify a user’s signature that converts a subscriber to a site based on their behavior over time.
Say my site has 6 different functions that can be used before subscribing, and users can convert a subscriber at any time.
For this user, I have statistics that represent the intensity in the continuous range of this interaction with functions 1-6 daily, therefore:
- D1: f1, f2, f3, f4, f5, f6
- D2: f1, f2, f3, f4, f5, f6
- D3: f1, f2, f3, f4, f5, f6
- D4: f1, f2, f3, f4, f5, f6
Say the user converts on day 5.
Which machine that uses the algorithms will help me determine which of the most common patterns in the use of functions lead to conversion?
(I know this is a super basic classification question, but I could not find a good example using longitudinal data, where the input vectors are ordered by time, like mine)
To continue this problem, suppose that each function has 3 intensities at which the user can interact (H, M, L).
Then we can represent each user as a string of states of interaction intensity. So for the user:
It would mean that on the first day they only interacted significantly with functions 5 and 6, but by the third day they very strongly interacted with functions from 3 to 6.
N-gram style
. (, "" )
"" n- , , , . , n-, . , n- , 10 n- ( ), , , .
Amaç Herdağdelen n- n, , n- . ( ). , n-, . ( 729 , , !)
, , n-, , , , ?
, , , -, , , . , Cox Proportional Hazard , , ( , ) - , , ,
, , . , , ? , .
, , , , .