Abstract: In order to solve the problems that the safety investment was insufficient and unreasonable in domestic coal mine enterprises, a safety investment model based on integrated support vector machine (SVM) and continuous ant colony algorithm (CACA) was proposed. The investment of coal mine safety was divided into five items, namely industrial health investment, safety technology investment, safety management investment, safety education investment and labor protection supplies investment, the nonlinear mapping between safety investment and safety guarantee degree was established by SVM, CACA iteration was used to find the optimal security input scheme under the premise of ensuring a certain degree of security. The results show that the scheme can optimize the allocation of security investment funds, avoid waste and inadequate investment and other problems. It proves the feasibility of the application of the safety investment model that integrated SVM and CACA, which can be further studied to guide coal mine enterprises to make scientific and reasonable safety investment.