Feature Selection - Optimality Criteria

Optimality Criteria

There are a variety of optimality criteria that can be used for controlling feature selection. The oldest are Mallows' Cp statistic and Akaike information criterion (AIC). These add variables if the t-statistic is bigger than .

Other criteria are Bayesian information criterion (BIC) which uses, minimum description length (MDL) which asymptotically uses, Bonnferroni / RIC which use, maximum dependency feature selection, and a variety of new criteria that are motivated by false discovery rate (FDR) which use something close to .

Read more about this topic:  Feature Selection

Famous quotes containing the word criteria:

    There are ... two minimum conditions necessary and sufficient for the existence of a legal system. On the one hand those rules of behavior which are valid according to the system’s ultimate criteria of validity must be generally obeyed, and on the other hand, its rules of recognition specifying the criteria of legal validity and its rules of change and adjudication must be effectively accepted as common public standards of official behavior by its officials.
    —H.L.A. (Herbert Lionel Adolphus)