The invention relates to a psychological behavior analysis method based on a facultative rate and FP-growth and belongs to the technical field of data mining. The method includes, firstly, establishing an affair database of human psychological behavior vocabulary data; then, solving a compatible storage rate and a minimum compatible storage rate of each psychological behavior vocabulary compatiblestorage rate item so as to reserve a strong association item and generate a new transaction set D; then, scanning the D for the first time, and calculating and reserving frequent vocabularies; then scanning D for the second time, and constructing a frequent pattern tree FP-tree, and extracting a frequent item set; secondly, generating a corresponding strong association rule by each frequent itemset, and calculating the confidence degree of each strong association rule; and finally, sorting the frequent item sets according to the confidence degree of each strong association rule, and visuallydisplaying the association degree of each psychological behavior vocabulary according to the result. According to the invention, the optimized PE-growth algorithm plays a role in correlation analysisof human psychological behavior vocabularies, and a more accurate and efficient analysis method is provided for disciplines such as psychology and micro-expressions.