The phrase "refine classification" is correct and usable in written English.
You can use it any time you want to refer to the process of developing and refining a set of categories or classes of similar items or ideas. For example, "Researchers are attempting to refine the classification system used to categorize types of dinosaur fossils."
Active learning can also be conducted to refine classification labels assigned to each sentence in the corpus.
MFS feature selection technique was combined with computerised feature selection process to further refine classification accuracies obtained in previous iterations.
Given the (i) multi-domain nature of many protein kinases, (ii) difficulty of domain assignment, and (iii) existence of domain shuffling, we wished to propose a protocol to refine classification that would: take into account full-length sequences, be insensitive to domain detection inaccuracies by avoiding the step of domain detection, take into account domain shuffling.
Full sequencing of breast cancer genomes could potentially refine classification and give a more complete picture of the mutational profile of cancer and thus aid therapy decisions.
Finally, we suggest that classical taxonomy and modern genetic approaches must work in tandem, not only to improve the accuracy of species identification but also to potentially refine classifications at the levels of organism, population and genotype in the field and laboratory.
Refining Classification of Pancreatic Cancer Subtypes to Improve Clinical Care.
It implies the usage of LVH over other traditional methods to treat multi-class data sets refining classification.
The resulting phylogenetic tree agrees very well with the recently refined classification of the major fungal lineages (Fig. 3).
As a consequence, refined classification of transcription factors based on nuclear stoichiometric abundance, association-dissociation kinetics, co-regulating partners, and the type of DNA they are bound to (methylated, condensed) will emerge and will challenge the existing classification [ 52, 53].
We did not test whether the performance of the rule-based model would be improved if we refined classification rules using statistics from the UCI data because these data were limited to only three subjects who were indoors 87% of the time and had non-representative outdoor walking data.
Over the past decade, as magnetic resonance imaging has refined classifications of disc degeneration [ 5, 92], it has become evident that, although factors such as occupation, psychosocial factors, benefit payments and environment are linked to disabling back pain [ 93, 94], contrary to previous assumptions these factors have little influence on the pattern of disc degeneration itself [ 95, 96].
