Aiming at the related data of unknown terrorist attacks, this paper proposes a method of mining antiterrorism information based on clustering optimization, which provides important information for the fight against terrorism. Firstly, principal component analysis is used to extract the main features of unknown attack events. The "Elbow Method" is used to select the cluster cluster number to determine the unknown attack events based on KMeans clustering algorithm. Then through nonlinear programming and the clustering optimization algorithm ,the terrorist organization's base prediction problem is transformed into an unconstrained optimization problem, and then the location of the terrorist organization is accurately estimated. The similarity between the typical event and the suspect is obtained,the simulation experiment is used to infer the position of the ISIL organization in the terrorist attack in Iraq in recent years. The results show that the method has certain significance and value for early warning of terrorist attacks.