predictions

evaluation of the model

estimating relations between object features and assigned label

forecasting

association

clustering

dimensionality reduction

optimization

physical connection from units in a later layer to units in a previous layer of a neural network

method of training neural networks to compute the network parameter updates

method of preselecting features based on the expected results

an assumption that the variance of the errors should be constant

an assumption that the independent variable should be error-free

one of the assumptions of Ordinary Least Squares regression

that the absolute values of residuals does not vary much

there are no duplications of the data

constant variance of errors in the dataset

independent variable is not contaminated with measurement errors

the process of assigning objects to pre-existing classes

the process of grouping objects into groups with similar features

the process of defining classes for new objects