Quiz 3
No description yet. This curated deck is ready for daily study.
Card preview (19)
Decision trees are an algorithm for which machine learning task?
Classification
Root node
decision tree has no incoming edges and zero or more outgoing edges
Internal node
has more than one incoming edge and one or more outgoing edges
Terminal node
has exactly one incoming edge and no outgoing edge
Internal node
denotes a test on a feature
In classification, a model or classifier
constructed to predict class (categorical) labels
Homogeneous class distribution
preferred to determine the best split for decision trees
Methods to look at node Impurity
Gini Index Information Gain Gain Ratio Misclassification Error
What is the minimum and maximum value for entropy?
0,1
What is the minimum and maximum value of GINI?
0,0.5
What is the minimum and maximum value for Misclassification Error?
0,0.5
The process of building a decision tree is recursive. It is built one node at a time and the algorithm is applied at each node the same way.
True
During the process of building a decision tree, at each node, the algorithm decides which attribute to use for a new split based on some criteria evaluated for each possible attribute.
True
During the process of building a decision tree, at each node, we consider the similarities between all the points and take the most similar ones to create the next nodes based on a criteria evaluated at each similar group of data.
False
What are splitting criteria used in decision trees?
GINI index Information Gain - Entropy Misclassification Error Gain Ratio
Building a tree that is too large or complex can cause
overfitting
To reduce the size of a tree after it is built, use an algorithm for
pruning
A decision tree tells how to classify instances. It can be turned into a list of rules. Which of the following are true?
Each leaf node gets turned into one rule Each internal node after the root gets added to the rule as an and condition
Overfitting in Machine Learning is defined as when a statistical model describes random error or noise instead of underlying relationship or when a model is excessively complex.
True