In the SVM classification, we use planes to classify the labels points if the dataset has 3 input features. We need to use planes when input features are 3. I am describing a toy dataset with 3 input features as follows

Study_time    rest_time   pass_time    label
40            10          5            Good
38            12          3            Good
20            8           10           bad
15            12          2            bad

In this dataset you can see, three input features are study_time, rest_time, pass_time. We need to define a plane to find out the label. I went through various course materials of support vector machines and every material said that we need to define points in 3D space if the number of features is 3.

I know points mean a dot which has an x co-ordinate and a y-co-ordinate.

In toy dataset, every instance has 3 values, 1 for study times, 1 for rest_time, 1 for pass_time. 1st instance can be defined as (40, 10, 5).

If we consider the points with respect to the toy dataset, which are those points? what are there co-ordinates?

Thank you.


1 Answer 1


By points they mean x,y and z components. Not just an x and a y. In the same way that you need two points to uniquely define a line, you need 3 "points" to define a plane. The points in the dataset are simply the numerical values for each row: (40,10,5),..., (15,12,2).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.