decision_trees.ipynb [Part 2] 1. Pima Indian Diabetes Dataset The Pima Indians Diabetes Data Set was developed by the United States National Institute of Diabetes and Digestive and Kidney Diseases.   Astonishingly, over 30% of Pima people develop diabetes. In contrast, the diabetes rate in the United States is 8.3% and in China it is 4.2%. Each instance in the dataset represents information about a Pima woman over the age of 21 and belonged to one of two classes: a person who developed diabetes within five years, or a person that did not. There are eight attributes in addition to the column representing whether or not they developed diabetes:

decision_trees.ipynb [Part 2]

1. Pima Indian Diabetes Dataset

The Pima Indians Diabetes Data Set was developed by the United States National Institute of Diabetes and Digestive and Kidney Diseases.

Astonishingly, over 30% of Pima people develop diabetes. In contrast, the diabetes rate in the United States is 8.3% and in China it is 4.2%.

Each instance in the dataset represents information about a Pima woman over the age of 21 and belonged to one of two classes: a person who developed diabetes within five years, or a person that did not. There are eight attributes in addition to the column representing whether or not they developed diabetes:

1. The number of times the woman was pregnant
2. Plasma glucose concentration a 2 hours in an oral glucose tolerance test
3. Diastolic blood pressure (mm Hg)
4. Triceps skin fold thickness (mm)
5. 2-Hour serum insulin (mu U/ml)
6. Body mass index (weight in kg/(height in m)^2)
7. Diabetes pedigree function
8. Age
9. Whether they got diabetes or not (0 = no, 1 = yes)

We are trying to predict whether they got diabetes or not based on the features.

The csv file at is at

https://raw.githubusercontent.com/yew1eb/machine-learning/master/Naive-bayes/pima-indians-diabetes.data.csv

This file does not have a header row

You will need to

1. load the file into a dataframe
2. divide the data into training and test sets. (an 80-20 split sounds good)
3. train a decision tree classifier on the training data
4. display the tree
5. run the classifier on the test data
6. compute the accuracy
7. Have a small paragraph describing the results.

Good luck!

[].....

[].....

[]....

2. The Wisconsin Cancer Datasett

The task is to predict whether a tumor is malignant or benign (the second column of the dataset based on 30 real values.

The data file is

https://raw.githubusercontent.com/zacharski/ml-class/master/data/wdbc.data

And a writeup about the data is at:

https://raw.githubusercontent.com/zacharski/ml-class/master/data/wdbc.names

Follow the same steps as above.

[]......

[].....