Although some network parameters including the number of layers or hidden neurons were selected in a manner to achieve the best performance of classification or estimation, no organized and comprehensive framework was proposed by the previous authors for selection of the best architectures and parameters of ANNs. 4.
Pharmacophore model Hypo1/7 illustrated the best combination of classification power (ROC-AUC) and correlation with bioactivity.
Table 4 presents the confusion matrix corresponding to the best result of classification.
We find that the Zernike polynomials provide better feature representation than the pseudo-Zernikes and that the decision trees yield the best balance of classification accuracy and interpretability.
Even the HF and LF gives the best performance of classification accuracy [7], which is approximately 88.33 and 79.27 %, respectively, but this classification accuracy is still not good enough to achieve better classification accuracy rate over 90%[11]1].
Table 12 summarizes the best results of classification on the dataset.
For the best result of classification it is therefore of crucial importance that all particle images be aligned to each other as precisely as possible.
We obtained the best possible accuracy of classification with an AUC of 0.84 and maximal accuracy of 77.1% by using a filtering distance of 10 kb for determining the set of distal constitutive exon intron boundaries in IMR90.
In overall, the location and multinomial gating functions seemed to perform the best in terms of classification performance and specificity/sensitivity when including all clinical variables, whereas the independent model seemed to be competitive once the clinical variables were selected on breast and CNS.
Meanwhile, paper [5] used wavelet transform for ECG signal decomposition and feature extraction; totally, six statistical features in combinations of LF, HF, and LF/HF give a best of 96.41 % classification accuracy rate.
Comparison of the three classification algorithms (NB, BN and NN) yielded a best algorithm in terms of classification accuracy.
