Association Rule is one of the main technologies of Data Mining, this paper combining with the particularity of network security, applying the association rule to large scale networks of situational awareness. Against the limitation of the classical algorithm for mining association rules Apriori algorithm, proposes an efficient algorithm to save time consumption. According to the three levels of reducing the number of non-frequent sets, avoiding calculating the items that is not existed and reducing the database redundancy items to improve the Apriori algorithm, saving lots of time and space in the algorithm process. Experiments results show that this algorithm can quickly and effectively mining association rules, and help to identify potential, malicious attacks, safeguard the overall network security situation.