Drug Propagation Prediction Model based on CA-SIS and SVM

In recent years, opioids have been widely used in the field of analgesia treatment, and they are irreplaceable in some aspects. However, when opioids are effective, they can suppress the human immune system to some extent and are addictive. The overuse of opioids will seriously affect our social stability. In the behavior of transmission model, the dynamic model of many nodes are based on average method, the differential equation based on the mean field theory to analyze propagation behavior, it is difficult to reflect the status and evolution of each node in the network situation, cellular automata can effectively solve this problem, so we use cellular automata as a simulation system, transmission rules, based on the SIS model combining both drug state propagation prediction model is established. Considering that the traditional SIS model is unable to evaluate the degree of propagation (the number of drug propagation) of each node, we used the Support Vector Regression (SVR) model to predict the number of propagation of each node in the network. SVR model has the ability to deal with nonlinear multi-dimensional small samples, which has a good effect on the nonlinear regression prediction of time series, and provides a new choice for the prediction of opioid prevalence.