Event Extraction is an important part of information extraction, as the basis of Event Extraction, Event Recognition directly affects the result of Event Extraction. The frequent subtree of dependency syntax tree (FSDST), represents the structure that the trigger word of event commonly used with. First We Mining FSDST, and then gain the rules of Recognition event, and add these rules in the programs of Event Recognition, at last recognize events. The experiments show Event Recognition integrating the rules of FSDST improves F-measure to 81.4%.