Decision Trees
Decision tree is one of the learning algorithm for classification or regression problems
Basically its a set of if then... else...
rules for identifying a solution.
-
Lecture start by giving various attributes or examples for going to a restaurant
-
Then it deeply explains on the concept using the golf playing on saturday decision
-
Best algorithm is ID3 for decision tree
-
Other concepts like Entropy, Information gain are used to identify the best root node based on which the decision tree is build
-
Cross validation can be done using testing the algorithm against the training data set using various combination of the models.
-
When few cases does not have any proper data either the average of other similar data or same data as others are considered in order to identify the node