However, most of the foundations of Data mining sit over the assumption of a large, complete enough dataset to substantiate accurate models, assuming stability on the "concepts" to be modelled.
These were enough dataset to be applied to small scale case studies (the buildings with the similar architectural style, such as brick exterior walls (Dimitrov and Golparvar-Fard 2014) and progress detection of concrete structures (Han and Golparvar-Fard 2015)).
For MR acquisition, there are not yet enough datasets available to systematically examine how reliability is affected by the particular acquisition protocols, although the current data suggest that sequences previously optimized for brain morphology analyses (i.e., those used in GSP dataset) will have better reliability.
Synthesizing both attack and normal data might not build robust enough datasets in that they are not similar enough to the real world.
Provided that large enough datasets of different tissues are available, this method can easily be applied to other microarray platforms.
One of the biggest challenges in developing valid prognostic genomic based biomarkers for breast cancer is obtaining large enough datasets with sufficient patient follow-up time [ 5, 6].
DACTAL has not been tested on datasets with fewer than 1000 sequences, where we anticipate little advantage in using DACTAL; indeed, it is likely that DACTAL will be less accurate than good two-phase methods on small enough datasets.
We then performed linear regression analyses if enough datasets were available in a group (n > 5), with the proportion of accurate grading per study as the independent variable and 1/√ ESS as the dependent variable.
Furthermore, some large datasets (such as 16S.T) can be fairly accurately aligned using the best alignment methods, and in these cases, DACTAL may also not provide any improvement in topological accuracy (although it may provide a running time advantage for large enough datasets).
What's more, to do anything really interesting, sheer brainpower won't be enough; you'll need massive datasets, to train your system.
