The experimental results demonstrate the ability of the method to identify possible segmentations in a dataset, compared to algorithms that only yield a single clustering solution.
Our experiments with three real-world datasets show that our approach gives an improvement of 50% to 300% in the crowd cost to resolve a dataset, compared to using only tasks of the same interface.
For upsampling a target size of 200% (number of instances) of the upsampled training dataset compared to the original dataset is employed.
Before we move on, we highlight two advantages of using this dataset, compared to surveys which collect data at lower (i.e. yearly or longer) frequencies.
In few cases, such as SP-pLSA shows slightly better results (83.7%) considering color SIFT for Scene 15 dataset compared to DTCTH (83.63%).
Using PCA as feature selection techniques and with the consideration of high AUC and PD values as well as low PF value, random forest and Naïve Bayes perform well for CM1 dataset compared to the others.
As described above, the incorporation of fpocket cavity detection into BSC introduces the potential for noise in the binding site dataset compared to only defining binding sites surrounding observed bound ligands, and this may result in poorer retrieval performance metrics.
Although, wikipedia-medium is a larger dataset compared to lshtc-large dataset, the L-BFGS algorithm converges in a shorter time, because it requires a smaller number of iterations to reach the optimum.
The results of ROC area of all the compared methods on the DOVES dataset are presented in the fourth column of Table 2. Method MR shows the highest ROC area (0.7375) on this dataset compared to the previous methods.
Although, the disadvantage of small training datasets was compensated to a degree by increasing the relative amount of active compounds in the training dataset compared to the test dataset.
This is due to the fact that there were more images in the two-class dataset compared to the ten classes' case (when some of the ten classes were not represented by enough images in the dataset).
