ほくそ笑む

R言語と統計解析について

Laplace accuracy

Classification based on predictive association rules (CPAR) is a more advanced AR based classifier based on information metric [34]. In CPAR, Laplace accuracy is used to measure the accuracy of rules. Given a rule r, it is defined as follows:

 \displaystyle \ \ Laplace \ accuracy\ (r) = \frac{(N_c + 1)}{(N_{total} + m)}

where  m is the number of classes,  N_{total} is the total number of examples that satisfy the rule's body, among which  N_c examples belong to the predicted class,  c of the rule. For classification, the best  k rules of each class are selected from the rule sets of each class. By comparing the averaged Laplace accuracy of the best  k rules of each class, the class with the best accuracy is chosen as the predicted class.

Rule Extraction from Support Vector Machines (Studies in Computational Intelligence)

Rule Extraction from Support Vector Machines (Studies in Computational Intelligence)