|
Title:
|
DETECTING SYSTEM INTRUSIONS BY USING BOTH LABELED AND UNLABELED DATA |
|
Author(s):
|
Eric P. Jiang |
|
ISBN:
|
978-989-8533-62-3 |
|
Editors:
|
Miguel Baptista Nunes, Pedro Isaías and Philip Powell |
|
Year:
|
2017 |
|
Edition:
|
Single |
|
Keywords:
|
Data Mining, Intrusion Detection System, Data Preprocessing, Semi-Supervised Learning |
|
Type:
|
Short Paper |
|
First Page:
|
185 |
|
Last Page:
|
189 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
In this paper, we propose a semi-supervised learning approach, which is based on the well-known AdaBoost algorithm, for system intrusion detection. The approach uses only a small set of labeled training data to build up initial models of normal and anomalous system activity behaviors, and then it applies additional unlabeled audit data to further refine the behavior models. Experiments with the approach on a variant of the KDD Cup 99 data have shown that the proposed semi-supervised approach delivers a high detection rate while maintaining a very low false positive rate, and it represents a viable and competitive method for detecting system intrusions. |
|
|
|
|
|
|
