The Iterative method is our baseline approach which is commonly utilized by existing tools [ 17, 18].
Unlike the baseline method, our approach performs almost equally well at all three imaging qualities.
In Jonnalagadda et al.'s experiment, the baseline method had an F-measure of 80.3%; while our baseline method achieved a much higher F-measure of 85.42%, which made it more difficult for further improvements.
For the implementation of ∝SVM, we choose alter- ∝SVM as our baseline method instead of conv- ∝SVM for the former performs better than the latter in most situations [11].
While this is good evidence of that densely connected biclusters indeed reliably reflect functional modules, we observed some further interesting phenomena: Our baseline method, which operates on interaction network data only, achieved respectable enrichment (ER) values (see Table 1 and 2), which underscores that connectivity is a valuable concept when screening interaction network data.
Therefore, in this study, our baseline method included all four types of features in [ 17] and the combined features in [ 28].
Among these, (Giuliano et al. 2006) and (Mitsumori et al. 2006) do not rely on natural language parsing, while the former applied SVMs with kernels on surface strings and the latter is similar to our baseline method.
The second dataset consisted of 600 SNV sites spanning a 1M bps genomic segment on human chromosome 10, which is a moderately challenging dataset whose re-identification risk can be fully mitigated by our baseline method.
The first dataset consisted of 311 SNV sites spanning of 5M bps genomic segment on human chromosome 2, which was a challenging dataset for which the re-identification risk could be alleviated (but not fully mitigated) using our baseline method (see below).
For benchmarking, we regard "STIP" and "iDT" features as our baseline methods and also made comparisons against other competing methods from literature.
