Delhi Delights!

Suppose you work at Delhi Delights! which is a food delivery company in Delhi. It offers a premium membership called $$Delighted Members$,$ with which there is no delivery cost for your order. Lately, the number of purchases of this premium membership has been going down. Now, based on past data, Delhi Delights! wants to predict which of the customers will buy the 'Delighted Members' membership and which ones will not. For this, you have been provided with the following dataset.

Delhi Delights Data

Download

Delhi Delights! Dataset $-$ The first two columns in the training set file represent the two attributes, Average Delivery Rating $($a$1)$ and Average Orders per Month $($a$2).$ The third column $($column C$)$ is the class label associated with each data point in the training set. There are two class labels: "Yes" and $"$No$"$ indicating whether the person has bought the premium membership. In total, there are $30$ data points$/$observations in the training set. You need to build a decision tree using this file. Use all $30$ points for training.

Please note that you mustn't use DecisionTreeClassifier$()$ or write codes to solve the following questions. You can use Excel to split and filter the data set according to the test conditions specified in the questions. The DecisionTreeClassifier function uses a slightly different test on each node, and hence, the final answer to the following questions may not match.

Please use the following hyperparameters to solve the following questions:

max$\_$depth $=5$

min$\_$samples$\_$split $=10$

min$\_$samples$\_$leaf $=5$

Homogeneity measure $=$ Gini

Note: $(13,5),$ written on a node, implies that $13$ data points belong to class label $"$No$",$ and $5$ data points belong to class label "Yes". This syntax is true for all the nodes shown in the tree.