Intrusion Detection Systems (IDSs) have turned into an important security system for managing dangers and a needful aspect of global security architecture.The IDS is triggering alerts for any suspicious activity which means thousand alerts that the analysts should take care of it. These Alerts has contained irrelevant and redundant features and most of them don't require big attention by researches. Deleting the alert attributes or reducing the amount of them from the entire amount alert attributes lead the researchers to create many methods such as principle component analysis. Feature ExtractionMining Method in IDS is an important data mining step.In this paper, we focus on an approach of feature selection based on Darpa 1999. The first step that based on data preprocessing and configuration for the next stage and guides the initialization of search process for the second step that based on principle component analysis whose outputs the final feature subset.