Most machine learning tools provide some parameters for users. For example, the k in knn classification algorithm and the k in k-means clustering algorithm. From the optimistic view, these parameters provide flexibility and make machine learning tools more powerful. However, from another point of view, these machine learning techniques cannot determine (or learn) some parameters automatically so that users must specify by themselves.
RVKDE features an average time complexity of O(nlogn) for carrying out the training process, where n is the number of instances in the training dataset. Here is a comparison against SVM in terms of the training dataset size.
Training time (in seconds) | Testing time (in seconds) | |||
---|---|---|---|---|
#samples | RVKDE | SVM | RVKDE | SVM |
50 | 29.6 | 138.1 | 54.5 | 146.7 |
100 | 91.7 | 527.0 | 87.6 | 301.0 |
250 | 486.4 | 5105.6 | 153.3 | 758.5 |
500 | 1377.4 | 21040.0 | 220.7 | 990.7 |
1000 | 3887.8 | 78795.3 | 333.2 | 2532.8 |
The table reveals that the training time with the SVM increases approximately in the order of O(n^2).