Advantages and Disadvantages of Rule-Based Classifiers
Advantages
Rule-based classifier has the following advantages.
1. Has characteristics quite similar to decision trees
a. As highly expressive as decision trees
b. Easy to interpret
C. Performance comparable to decision trees
d. Can handle redundant attributes benight
2. Better suited for handling imbalanced classes
Disadvantages
The rule-based classifier has the following disadvantages
1. Harder to handle missing values in the test set
2. If the ruleset is large then it is complex to apply the rule for classification.
3. For large training set the large number of rules generated requires a large amount of memory.
4. During rule generation extra computation is needed to simplify and purn the rules.
Comments
Post a Comment