7 Classification and Discrimination

Discriminate analysis and classification are multivariate techniques concerned with separating distinct facts of objects (observation) and with allocating new objects (observations) to previous define groups.

Discriminate analysis is rather exploratory while classification require well defined rules.

In classification clauses of misclassifications and costs of missclassifications are sometimes taken in account when developing rules.
A Bayesian approach is encompassed as the way of introducing prior information.

Some methods assure specific distribution sources as the normal distribution.

Questions on this section

Stuck on something here? Ask below and it stays attached to this topic.