Data Science Asked by Syntxa erorr on April 12, 2021
I have the Data of 10,000 users Time Session in a website/App, The Login time, logout time, the person activity,
The Data is available for 60 days ( per user )
Using this 60 days data for 10k users, can I predict the active time of the respective user on 61st day? if yes what is the best approach and please suggest which type of problems I can refer to solve this
Thank you
Convert your training data to the following format -
a = login_time
b = logout_time
c = person_activity
train_data_X = [
[[a, b, c], [a, b, c], ......, [a, b, c]]
[[a, b, c], [a, b, c], ......, [a, b, c]]
.
.
.
[[a, b, c], [a, b, c], ......, [a, b, c]]
]
train_data_Y = [
[[time_day_1], [time_day_2], ...... [time_day_60]]
.
.
.
[[time_day_1], [time_day_2], ...... [time_day_60]]
]
Explaination -
You Input(training_data_X) data has shape 10000 * 60 * 3
because you have 10K users, data for 60 days and each data point depends on 3 variables namely a, b and c.
Your Target(training_data_Y) data has shape 10000 * 60 * 1
because you have just one number to output i.e active time.
Now, this is a Regression Problem because you are trying to predict a continuous value i.e active time.
You can model this type of problem with a number of time-series models. But since you got a multivariate input, i.e [a, b, c](refer training_data_X)
. You would be benefitted a lot using a Seq2Seq model to generate your time-series. If just want to generate 61st time-step you could use it just for that as well.
Answered by ashutosh singh on April 12, 2021
Get help from others!
Recent Questions
Recent Answers
© 2024 TransWikia.com. All rights reserved. Sites we Love: PCI Database, UKBizDB, Menu Kuliner, Sharing RPP