Hello All,
I am Omm prakash sahoo, a 3rd student from IIT BHU and a Data Science BS 
student from IIT Madras.

Currently, we are working on a project, i.e., on Detecting Insider Treats 
from publicly available data *CERT insider threat dataset*.
Since the CERT insider threat dataset r4.2 contains more insider threat 
instances than other versions of the dataset, we choose it as the 
experimental dataset. The dataset captures 32,770,227 events (log lines) of 
1,000 users (with 70 insiders) over 17 months, which consists of five types 
of event records, including logon, device, email, file, and HTTP. Among 
these are 7,323 malicious events generated by 70 insiders, representing 
three insider threat scenarios (Information leak, Data theft, and IT 
sabotage).

Due to this highly unbalanced data distribution, we want to create a GAN 
model to augment the malicious data to a reasonable size. So, please help 
us if there is any seqGAN code available to solve the issue. 

Thank you,
Omm prakash








-- 
Datameet is a community of Data Science enthusiasts in India. Know more about 
us by visiting http://datameet.org
--- 
You received this message because you are subscribed to the Google Groups 
"datameet" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/datameet/42deb383-8c1b-4b3d-b56b-b38869da20can%40googlegroups.com.

Reply via email to