Figure 5 Example images from video sequences in KTH dataset (publicly available at ).http://www.nada.kth.se/cvap/actions/ This dataset was benchmarked with both unsupervised and supervised methods in the current study.
Using real datasets avoids simulation biases and gives a real picture of mapper behavior, whereas simulated datasets are benchmarks from which all parameters can be controlled.
The dataset has been benchmarked in the literature [ 10, 20, 21].
The use of a UniProt derived dataset has recently been benchmarked for a number of classifiers with satisfactory results [ 36].
In our numerical experiments, thirteen real microarray datasets were used as benchmark datasets and nine algorithms including KNN, SKNN, IKNN, LS, LLS, ILLS, SLLS, BPCA and SVD were used.
Time series, non-time series and mixed type datasets were used as benchmark datasets, and the performance of each algorithm was evaluated using different measures mentioned above.
The six microarray datasets were used as benchmark datasets in numerical experiments to compare the performances of our shrinkage regression-based methods and the original regression-based methods.
Note that the two benchmark datasets were modelled using neural network training, while we modelled the P-gp data with a decision tree method.
To validate the performance of FARO in a more quantitative fashion, two benchmarking datasets were created from the Rosetta compendium of yeast gene expression profiles [4].
To further assess the performance of SRF, a total of 38 multiclass benchmark datasets were used.
