Abstract
LEM2 algorithm, a rule induction algorithm used by LERS, accepts input data sets only with symbolic attributes. MLEM2, a new algorithm, extends LEM2 capabilities by inducing rules from data with both symbolic and numerical attributes including data with missing attribute values. MLEM2 accuracy is comparable with accuracy of LEM2 inducing rules from pre-discretized data sets. However, compared with other members of the LEM2 family, MLEM2 produces the smallest number of rules from the same data. In the current implementation of MLEM2 reduction of the number of rule conditions is not included, thus another member of the LEM2 family, namely MODLEM based on entropy, induces smaller number of conditions than MLEM2.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
