Naive Bayes modification for intrusion detection system classification with zero probability

One of the methods used in detecting intrusion detection system is by implementing Naïve Bayes algorithm. However, has a problem when one probabilities 0, it will cause inaccurate prediction, or even no prediction was found. This paper proposed two modifications for The first modification eliminated variable that 0 probability and second changed multiplication operations to addition operations. only applied algorithm does not find any results caused zero probabilities. this research show value precision, recall, accuracy made tends increase better than original highest are obtained from changing operation addition. Increasing precision can reach 4%, increasing recall reaches 2% 2%.