Exploratory-data-analysis

The dataset contains cases from a study that was conducted between 1958 and 1970 at the University of Chicago’s Billings Hospital on the survival of patients who had undergone surgery for breast cancer.

Description

The dataset contains cases from a study that was conducted between 1958 and 1970 at the University of Chicago’s Billings Hospital on the survival of patients who had undergone surgery for breast cancer.

Attribute Information

Column 1: Age of patient at the time of operation (numerical)
Column 2: Patient's year of operation (year 1900 - numerical)
Column 3: Number of positive axillary nodes detected (numerical)
Column 4: Survival status (class attribute)
1 = the patient survived 5 years or longer
2 = the patient died within 5 years

Loading the dataset

"""

import pandas as pd import seaborn as sns import matplotlib.pyplot as plt import numpy as np plt.rcParams.update({'font.size': 16})

'''IMPORTANT''' '''download haberman.csv from the classroom and upload the dataset in the content file in the left pane of the colab''' #Load haberman.csv into a pandas dataFrame. data = pd.read_csv("haberman.csv")

"""## PART 1: Understanding the dataset

Ques 1.1: Print the first 5 entries from the csv file

Expected Output

"""

write your code here

data.head(5)

"""### Ques 1.2: How many data-points and features does the dataset have?

Expected Output :

"""

write your code here

data.shape

"""### Ques 1.3: What are the column names in our dataset?

Expected Output:

"""

write your code here

data.columns

"""### Ques 1.4: How many data-points for each class of column 'status' are present?

Expected Output:

"""

write your code here

data['status'].value_counts(dropna= False)

"""## PART 2: Scatter Plot (2-D)

Just run this code cell (Don't change anything)

"""

Scatter Plot without color grading

data.plot(kind='scatter', x='age', y='year'); plt.show()

OBSERVATION: Both the classes are of same color (blue) and thus cannot make much sense.

"""### Ques 2.1: Color the points by their class-label/status-type

Expected Output:

"""

Write your code here

'''HINT: Use hue dimension of the Grid''' sns.set_style("whitegrid") sns.FacetGrid(data, hue='status', height=4)
.map(plt.scatter, 'age', 'year')
.add_legend() plt.show();

"""### Ques 2.2: Make 2 separate plots each of only 1 status-type in column dimension of the grid

Expected Output:

"""

write your code here

sns.set_style("whitegrid") sns.FacetGrid(data, col='status', height=4)
.map(plt.scatter, 'age', 'year')
.add_legend() plt.show();

"""## PART 3: Pair Plots

Ques 3.1: Plot the pairwise scatter plot with hue dimension of the grid

Expected Output:

"""

write your code here

plt.close(); sns.set_style('whitegrid') sns.pairplot(data, hue='status', height=5); plt.show()

"""## PART 4: PDF and CDF

Ques 4.1: Show the density plot with respect to the 'year' feature in 'row dimension' of the grid.

Expected Output:

"""

write your code here

sns.set_style("whitegrid") sns.FacetGrid(data, col='status', height=5)
.map(sns.distplot, 'year')
.add_legend() plt.show();

"""### Ques 4.2: Plot the CDF and PDF of 'age' of patients who died within 5 years.

Expected Output:

"""

write your code here

df = data[data['status'] == 2] counts, bin_edges = np.histogram(df['age'], bins=10,density=True) pdf = counts/sum(counts) cdf = np.cumsum(pdf) plt.plot(bin_edges[1:], pdf) plt.plot(bin_edges[1:], cdf) plt.show()

"""### Ques 4.3: What percentage of people who died were less than 55 years old.

Expected Output:

"""

write your code here

df = data[data['status'] == 2] df1 = df[df['age'] < 55] a = df.shape[0] b = df1.shape[0] c = round(b/a*100) print(f'percentage_dead_under_55 = {c}')