The standard DECISION-TREE-LEARNING algorithm described in the chapterdoes not handle cases in which some examples have missing attribute values.a. First, we need to find a way to classify such examples, given a decision treethat includes tests on the attributes for which values can be missing. Supposethat an example x has a missing value for attribute A and that the decisiontree tests for A at a node that x reaches. One way to handle this case isto pretend that the example has all possible values for the attribute, but toweight each value according to its frequency among all of the examples thatreach that node in the decision tree. The classification algorithm should followall branches at any node for which a value is missing and should multiply theweights along each path. Write a modified classification algorithm for decision trees that has this behavior.b. Now modify the information-gain calculation so that in any given collectionof examples C at a given node in the tree during the construction process,the examples with missing values for any of the remaining attributes are given”as-if ” values according to the frequencies of those values in the set C.
The post The standard DECISION-TREE-LEARNING algorithm described in the chapterdoes not appeared first on Get Papers.
Source: My posts