Representativeness-Based Instance Selection for Intrusion Detection

With the continuous development of network technology, an intrusion detection system needs to face efficiency and storage requirement when dealing with large data. A reasonable way alleviating this problem is instance selection, which can reduce space improve by selecting representative instances. An not only in its class but also different classes. This representativeness reflects importance instance. Since existing selection algorithm does take into account above situations, some selected instances are redundant important removed, increasing reducing efficiency. Therefore, a new proposed considers influence all same on classes Moreover, it as advantageous factor. Based representativeness, two algorithms handle balanced imbalanced data problems for detection. One representative-based data, named RBIS selects proportion from each class. The other RBIS-IM majority according number minority Compared benchmark sets detection, experimental results verify effectiveness demonstrate that achieve better balance between accuracy reduction rate or rate.