n$-$class Exercise: Classification

You$$ve been hired by Amazon to work on their recommendation system for books. Your

first task is to see if you can predict the recommendations of a single user, using the

following two attributes: Genre $($Romance$,$ Self$-$Help, or Thriller$)$ and Price $($High$,$

Medium, or Low$).$ The classification is whether the user recommended the book or

not.

Qa: Suppose you use decision tree algorithm and get the following classifier.

How would you classify a new instance as follows?

Book Genre Price Class

B$7$ Self$-$Help Medium

Qb: Use the following test data set to evaluate the decision tree classifier.

Test data set

Book Genre Price True Class

B$1$ Romance Low Not Recommended

B$2$ Romance Medium Recommended

B$3$ Thriller Low Recommended

B$4$ Thriller Low Not Recommended

B$5$ Self$-$Help High Not Recommended

B$6$ Romance High Recommended

If the classification outcome is recommended, we consider it as a positive. If the

classification outcome is not recommended, we consider it as a negative. Based on the

test data set and the decision tree classifier, derive the confusion matrix.

Confusion Matrix

True Positive$=?$ False Positive$=?$

False Negative$=?$ True Negative$=?$

Suppose you get the following confusion matrix based on a classifier. what is the

accuracy, precision, and recall of this classifier? How to understand these measures?

Confusion Matrix

True Positive$=40$ False Positive$=30$

False Negative$=10$ True Negative$=60$Qa: Suppose you use decision tree algorithm and get the following classifier.