INTERPRETING CHANCE FOR COMPUTER SECURITY BY VITERBI ALGORITHM WITH EDIT DISTANCE

This paper addresses the importance of chance discovery in computer security. There are various methods to discover chances in computer usage, but they have such drawbacks as discovering only anomalies, not interpreting anomalies in conventional approach in the community of computer security. This paper focuses on a role of interpreting the type of anomalies by analyzing the state sequences using Viterbi algorithm and evaluating the distance between the standard model of anomaly type and the state sequence of discovered anomalies. Because the state sequences are not always extracted consistently due to environmental factor, the edit distance is utilized to measure the distance effectively. Experimental results show the possibility of the proposed method in computer security.