The invention provides a pet feeding method and system based on big data analysis of Internet of Things. The method comprises the steps that the kind, gender, age, heartbeat frequency, blood pressure, body temperature, activity, feeding type and feeding amount of a pet are acquired, a current image and current weight form an influence factor matrix X and are uploaded to a server, wherein the feeding type and the feeding amount form decision variables; an Elman neural network is utilized to establish the complicated nonlinear relation between the influence factor matrix X and pet health indexes in a server, and a pet feeding model is obtained; an SPEA-II algorithm is utilized to optimize the pet feeding model, and a group of optimal solutions of the decision variables is obtained; the group of optimal solutions of the decision variables serves as a recommended decision X* of the pet to be issued to a terminal device of a user through the server for display; the user feeds the pet according to the recommended decision X* displayed by the terminal device. By utilizing the pet feeding method and the pet feeding system, an optimal pet feeding scheme can be determined, and a good living environment is created for the pet.