I have a slight misunderstanding as to what exactly is a one - dimensional SVM as mentioned by this paper: Training SVMs in 1D by Y-Su
Kindly correct me if my understanding of 1D SVMs is wrong. From my understanding of the paper, I felt that 1D SVMs are basically SVM's that operate on single dimension of the data. However another hypothesis lies in my head that 1D SVM's maybe be defined by the dimension of the hyperplane.
Kindly let me know your understanding after going through the above mentioned paper.
Regards, Leroy
Based on the paper, (and also on algorithm 1 in this paper that cites the paper you provided) 1D means that all the data points lie in one dimensional space. That is, all of them lie on one line. For example, you can imagine that all of them lie on the line
y=0, or on the liney=2x.