Hi Tayfun,
As per my understanding you would like to predict a continuous target feature using continuous numeric features. A classification model works well for situations where the target feature is discrete in nature. A regression model on the other hands predicts a continuous target feature. PFB the code of a linear regression model that can be used to achieve your task.
mdl = fitlm ([1;2;3;4],[2;4;6;8]);
pred = predict(mdl,[5;6;7;8]);
For more information, refer to the following link.
