Haberman's Cancer Survival - Exploratory Data Analysis

 

Objective is to perform exploratory data analysis on "Haberman’s Cancer Survival data set" and note the observations on data in plain crispy language.

 

Perform a analysis on this data set with the following sections:

  1. High level statistics of the data set: number of points, number of features, number of classes, data-points per class.
  2. Perform Univaraite analysis(PDF, CDF, Boxplot, Voilin plots) to understand which features are useful towards classification.
  3. Perform Bi-variate analysis (scatter plots, pair-plots) to see if combinations of features are useful in classfication.
  4. Write your observations in english as crisply and unambigously as possible. Always quantify your results.

Click here for more details about data set.

 

Solution: (Python code below)

Observations:

  1. Number of data points are 305 and number of features are 4 as patient age, operation_year, axillary_node_count and survival_status
  2. Almost 73% of the patient survived beyond 5 years and almost 27% of the patients died within 5 years. 224 number of data points found for class, patients survived beyond 5 years. 81 number of data points found for class, patients died within 5 years . This is imbalanced data set
  3. Objective : Our main objective is to find out featues and conbibation of features which are useful towords classification
  4. Survival satus is only the feature useful towords classification
  5. No combination of feature found towards classification of data
  6. Quantifiable other observations are provided above below the plotted graphs.


Python Code Jupyter Notebook on Github - Click here







 

Previous Post Next Post