This document summarizes a presentation on using machine learning to diagnose breast cancer. It introduces machine learning and explains that it uses statistical techniques to allow computer systems to learn from data without being explicitly programmed. It then provides an overview of breast cancer, risk factors, and statistics. It states that machine learning will be used to analyze breast cancer biopsy data to make diagnoses. The document outlines the steps of collecting and exploring the biopsy data, preparing training and test datasets, training a k-nearest neighbors model on the data, and calculating the model's accuracy on the test data using a confusion matrix.

1. Breast Cancer Diagnosis
Mini Project Presentation for the MCA
Course by Ankit Gupta

2. Introduction
What is Machine Learning and why we are using it ?
Machine learning (ML) is a field of artificial intelligence that uses statistical
techniques to give computer systems the ability to "learn" (e.g., progressively
improve performance on a specific task) from data, without being explicitly
programmed.
ML
AI Self Thinking
Computer
Provides
Mathematical
tool to AI
11/23/2018Ankit Gupta 1719214832 2

3. Breast Cancer: An Overview
• Breast cancer is the second leading cause of cancer death in women, second only to lung
cancer.
• The leading risk factor for breast cancer is simply being a woman. Though breast cancer does
occur in men, the disease is 100 times more common in women.
• Men can also get breast cancer. In 2017, the American Cancer Society estimates 2,470 new
cases of invasive breast cancer will be diagnosed in men in the U.S.
• A woman has about a one in eight chance of being diagnosed with breast cancer in her
lifetime, according to the National Cancer Institute.
• Most women (about eight out of 10) who get breast cancer do not have a family history of the
disease.
• But women who have close blood relatives with breast cancer have a higher risk. Having a
first-degree relative (mother, sister or daughter) with breast cancer almost doubles a woman’s
risk.
Ankit Gupta 1719214832 11/23/2018 3

4. Why I am using Machine Learning?
Breast
Cancer
Biopsy
Data
Machine
Learning
Diagnosis
11/23/2018Ankit Gupta 1719214832 4

5. Understanding the Algorithm
Lazy Learning – Classification Using Nearest Neighbors
K-Nearest Neighbor classifiers are defined by their characteristic of classifying
unlabeled examples by assigning them the class of similar labeled.
Examples:
• Computer vision applications, including optical character recognition and facial
recognition in both still images and video.
• Predicting whether a person will enjoy a movie or music recommendation.
• Identifying patterns in genetic data, perhaps to use them in detecting specific
proteins or diseases
Ankit Gupta 1719214832 11/23/2018 5

6. Flow chart
• MeasurementsBiopsy
procedure
• Evaluation
Reports Diagnosis
Ankit Gupta 1719214832 11/23/2018 6
Analysis of
Measurements
Preparation of
ML Models
Predictions &
Validation

7. Take a View of Biopsy Data
Ankit Gupta 1719214832 11/23/2018 7

8. Analysis with R language
Step 1 – collecting data
We will utilize the Wisconsin Breast Cancer Diagnostic dataset from the UCI
Machine Learning Repository at http://archive.ics.uci.edu/ml. This data was
donated by researchers of the University of Wisconsin. The breast cancer data
includes 569 examples of cancer biopsies, each with 32 features.
Ankit Gupta 1719214832 11/23/2018 8

9. Kaggle Competition
Ankit Gupta 1719214832 11/23/2018 9

10. Step 2 – exploring and preparing the data
Importing the data set in R IDE
The table( ) output indicates that 357 masses are benign while 212 are malignant
Ankit Gupta 1719214832 11/23/2018 10

11. Visualization 1.
Ankit Gupta 1719214832 11/23/2018 11
Do you notice anything
problematic about the
values?

12. Visualization 2.
Ankit Gupta 1719214832 11/23/2018 12
So let's apply
normalization to rescale
the features to a
standard range of values

13. Visualization 3.
Ankit Gupta 1719214832 11/23/2018 13
A vector x of numeric values,
and for each value in x,
subtracts the
minimum value in x and
divides by the range of
values in x.

14. Data preparation – creating training and test
datasets
We will split the wbcd_n data frame into wbcd_train and wbcd_test:
If the preceding commands are confusing, remember that data is extracted from
dataframes using the [row, column] syntax. A blank value for the row or column
value indicates that all the rows or columns should be included. Hence, the first
line of code takes rows 1 to 469 and all columns, and the second line takes 100
rows from 470 to 569 and all columns.
Ankit Gupta 1719214832 11/23/2018 14

15. Step 3 – training a model on the data
Ankit Gupta 1719214832 11/23/2018 15
Now Shit back and relax all the work will be done by R language you
don’t have to do any more calculations. Unless you don’t know the
value of K in KNN Algorithm
k=sqrt(nrow(wbcd))
Square root of number of rows in the dataset & should be odd .

16. Confusion Matrix to calculate Accuracy
77 0
2 21
Ankit Gupta 1719214832 11/23/2018 16
Benign
Malignant

17. Accuracy
98.24%
Machine
Learning
Ankit Gupta 1719214832 11/23/2018 17

18. Ankit Gupta 1719214832 11/23/2018 18