IDAX.FMEASURE - Calculate the F-measure from a confusion matrix

Use this stored procedure to calculate the F-measure of a class from a confusion matrix. The F-measure, also known as F-score, is a balanced mean between precision and recall. It is calculated as 2 * TPR * PPV / (TPR + PPV), where TPR and PPV are the result of the corresponding stored procedures that are called on the same confusion matrix.

Authorization

The privileges held by the authorization ID of the statement must include the IDAX_USER role.

Syntax

IDAX.FMEASURE(in parameter_string varchar(32672))

Parameter descriptions

parameter_string

Mandatory one-string parameter that contains pairs of <parameter>=<value> entries that are separated by a comma.

Data type: VARCHAR(32672)

The following list shows the parameter values:

matrixTable: Mandatory.; The name of the confusion matrix table.; Data type: VARCHAR(256)
class: Mandatory.; The class in the confusion matrix table for which the F-measure is calculated.; Data type: VARCHAR(ANY)

Returned information

Double the F-measure as a result set.

Example

CALL IDAX.SPLIT_DATA('intable=samples.customer_churn,traintable=cust_train,testtable=cust_test,id=cust_id,fraction=0.30');
CALL IDAX.GROW_DECTREE('model=cust_dt, intable=cust_train, id=cust_id, target=censor, incolumn=duration:ignore;censor:nom, minsplit=2');
CALL IDAX.PREDICT_DECTREE('model=cust_dt, intable=cust_test, outtable=cust_dt_out');
call IDAX.CONFUSION_MATRIX('intable=cust_test, id=cust_id, target=censor, resulttable=cust_dt_out, matrixtable=cust_dt_cm');
call IDAX.FMEASURE('matrixTable=cust_dt_cm, class=0');