@extends("base.html") @require(htmlbase, view, pluginbase) @def content():
Data binning is a widely used data pre-processing technique that can be used to reduce the effects of minor observational errors. It can also be used to reduce the size of a dataset for easier processing. Regions of the data (bins) are defined, within which data is replaced with a representative value, e.g. a mean, median or central value. It is a form of quantization.
It is important to remember that binning results in a loss of data, and may introduce artefacts. This plugin attempts to mitigate some of these problems by providing information on data loss, and offering adaptive binning approaches.
This plugin can bin any data in 1 dimension, e.g. NMR spectra. Simply select the data source and the data will be automatically binned. You can adjust the bin size using the toolbar, together with a bin offset. You can view the result of the binning a tabular, or spectral view. Importantly, this plugin also presents a difference view that shows the data loss associated with a given binning strategy, together with a summary view.
Resulting data output can be fed into statistical analysis, for example PCA, where the effects of binning can be observed live.
@end