Nmap Development mailing list archives

Prabhjyot's Status Report #11 of 17


From: Prabhjyot Singh Sodhi <prabhjyotsingh95 () gmail com>
Date: Tue, 12 Jul 2016 10:12:01 +0530

Hey list,

As another week passed by, Here is all that I've accomplished

Accomplishments:
- Wrote and pushed some docs on nmap-private-dev/fp/ipv6/gsoc_trial, These
docs along with my last 3-4 status reports should help you understand how
we are going about things.

- Wrote code for getting testing the multi-stage random forest model. When
tested on the entire dataset, it is able to predict all of them correctly.
For 80:20 per group split, it is able to predict 51 correctly out of 68.

- Read about converting the makefile generated by cmake to autotools but
haven't been able to start work on it yet.

Priorities:
- Tune the model and make sure that we aren't overfitting the model. Given
the small size of our prints db it is very easy to create models that'll
end up performing badly once the dataset is larger.
- start working on the makefile conversion.
- Go through existing nmap code that relates to integrating model
predictions with final output.

Please find some time to go through documentation and free to suggest
changes or questions if anything is unclear.

Thanks,
Cheers,
Prabhjyot
_______________________________________________
Sent through the dev mailing list
https://nmap.org/mailman/listinfo/dev
Archived at http://seclists.org/nmap-dev/

Current thread: