Ordinal Logistic Regression allows comparison of predictors with respect to each other according to their p-value. Also, within each predictor, it is possible to quantify the influence of its variation on the outcome variable (i.e. category or class). For categorical predictors, you will be able to interpret the odds that one group had a higher or lower value on the outcome variable compared to another group. For continuous variables, you will be able to interpret how a single unit increase or decrease in that variable was associated with the odds of the outcome variable having a higher value. My question is whether from an SVM model, such information about the influence of predictor variables on the outcome variable can be acquired.