Book - Chapter 7 Analytical Theory Classification Flashcards
What applications does classification appear in
Data mining
What is the primary task of a classifier
To assign class labels to new observations
Are classification method supervised or unsupervised
Supervised
What is another name for a decision tree
Prediction tree
What is the input variable of a decision tree
Categorical or continuous
In a decision tree structure what is a test point
A node
What is a node without further branches called
A leaf node
What do leaf nodes return
They return class labels and, in some implementations, they return the probability scores
What are the two varieties of decision trees
Classification trees and regression trees
What are classification trees
They usually apply to output variables that are categorical for example often binary yes or no
What are regression trees
They can apply to output variables that are numerical continuous, such as the predicted price of a consumer good or the likely heard a subscription will be purchased
What does the term branch mean in decision trees
Refers to the outcome of a decision and is visualised as a line connecting two Nodes
What happens if the decision is numerical
The greater than branch is usually placed on the right
What is an internal node
Are the dissertation or test points. Each internal note refers to an input variable or an attribute
What is the top internal node called
The root
What is the depth of a node
Is the minimum number of steps required to reach the node from the root
What are short trees also known as
Weak learners or base learners
What’s on in ensemble Mefford
They use multiple predictive models to vote, and decisions can be made based on the combination of the votes
Gave examples of ensemble methods
Random forest, bagging, and boasting
What is the simplest short tree called
Decision stump
At each split what does the decision tree algorithm do
It picks the most informative attribute out of the remaining attributes
How is the most informative attribute determined
By measures such as entropy and information gain
What does entropy measure
The impurity of an attribute
What does information gain measure
The purity of an attribute
When do you achieve maximum entropy
When all class labels are equally probable
What is conditional entropy always
Less than or equal to the base Entropy
What is information gain defined as
The difference between base Entropy and conditional entropy
What is Bayes theorem
Gives a relationship between the probabilities of two events and their conditional probabilities
What is a naive Bayes classifier
Assumes that the presence or absence of a particular feature of a class is unrelated to the presence or absence of other features
What are the input variables of naive Bayes
Categorical and I’ll discreet
What is the output of naive Bayes
Class label and its corresponding probability score. The probability score is not the true probability of the class label, but it’s proportional to the true probability
What is naive Bayes most commonly used for
Spam filtering
What is Bayes theorem
The conditional probability of event C occurring, given that event A has already occurred, is to noted as P (C|A)
What should a good classifier have
A large true positive and true negative and a small (ideally zero) numbers for false positives and false negatives
What does accuracy mean
Defining the rate at which a model has classified the records correctly
What is recall
The percentage of positive instances that were correctly identified