As most of the pumps analyzed in this data set are ASHPs, we can estimate that majority of annual savings are somewhere in the range demonstrated in Table 1.
We compared the variation in bias metrics among several data sets: the two data sets analyzed in this paper, two Latin square spike-in data sets, and five tumor data sets that are relevant to our research interests [14] [18].
Thus, by this measure, the two data sets analyzed in this study are representative of "typical" clinical data sets, whereas the spike-in data sets are not.
We found that the colon and cell line data sets analyzed in this study are comparable to the tumor data sets, whereas the spike-in data sets exhibited much lower variation in the bias metrics (Figure 1).
The data set with severity measures was the main data set analyzed in this manuscript and we refer to it as the primary or first data set.
In the data sets analyzed in this paper, the magnitude of the eigenvalues corresponding to the set of eigenvectors decreases dramatically after the first few dimensions and the number of relevant dimensions for both MDA and PCA subspaces is comparable most of the time.
Of the test data sets analyzed in this work, CDPs identified the approved non-anticancer drugs as the set with the highest combined scaffold and MACCS keys/Tanimoto similarity.
The algorithm parameter defining the area constrain operation for guide-wire segmentation was set equals to 3500 pixels and kept constant for all the data sets analyzed in this study.
The data sets analyzed in this study are (i) the Thomson Reuters SDC Platinum database, listing around 15,000 inter-firm R&D alliances and (ii) a data set of over 300,000 co-authored papers in physics, which was obtained from the American Physical Society (APS) scholars database with additional disambiguation of authors names.
The data set analyzed in this study was obtained from ten ground-based magnetometer stations in Africa along the 96° MM [magnetic meridian] longitudinal sector.
