Apriori based Association Rule Mining (ARM) is one of the data mining techniques used to extract hidden knowledge from datasets that can be used by an organization’s decision makers to improve overall profit. Performing Existing association mining algorithms requires repeated passes over the entire database. Obviously, for large database, the role of input/output overhead in scanning the database is very significant. We propose a new algorithm, which would mine frequent item sets with vertical format. The new algorithm would need to scan database one time. And in the follow-up data mining process, it can get new frequent item sets through 'and operation' between item sets. The new algorithm needs less storage space, and can improve the efficiency of data mining.