Project Details
Safetica - Použití technik síťové analýzy v rámci prevence ztráty dat
Project Period: 1. 5. 2017 – 31. 12. 2017
Project Type: contract
Partner: Safetica
Detecting anomalies on a computer network
We offer you, in the framework of research, an analysis of the current state of
anomaly detection approaches on a computer network to use your data leakage
detection problem. Detecting anomalies is a long-term concern, and we have a set
of procedures and approaches to that.
The implementation itself
would be preceded by a preparatory phase in which we will get acquainted with the
data collection data collected and we will propose their adaptation for the
purposes of the subsequent analysis of the anomalies. Subsequently, we develop
a search for current knowledge of anomaly analysis on the network, with emphasis
on those that do not require the expert knowledge of the user (the customer using
the applicant's products), and we will propose candidates for examination in
a subsequent proof-of-concept practical test. The time-consuming phase of the
preparatory phase and the search is estimated at 120-160 hours and will be
realized during the first 3 months of the project.
As part of the
implementation of the proof-of-concept solution, we will verify the performance
of the proposed candidates (the solution principle) on the data you provide. The
evaluation of the approaches will be included in the final technical report. As
part of our collaboration, we also offer consultation on the problem and
consultation to take on the proposed solutions. Overall, the implimentation and
evaluation phase is estimated at 450-550 hours spread over 5-6 calendar months.
Holkovič Martin, Ing., Ph.D.
2017
- Final Report - Machine Learning Outlier Detectionin Safetica's Data Loss Prevention System, summary research report, 2017
Authors: PLUSKAL, J. - Machine Learning Outlier Detection in Safetica's Data Loss Prevention System, summary research report, 2017
Authors: PLUSKAL, J.