prajwal-144 / Cancer-Tumor-Prediction-SVM

SVM Model for predicting if the tumor is Malignant or Benign

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Cancer-Tumor-Prediction-SVM

SVM Model for predicting if the tumor is Malignant or Benign

SVM

Support Vector Machine is a supervised learning algorithm used mostly for classification in Machine Learning and Deep Learning. It looks at extremes of the dataset and draws a hypothetical decision boundary known as Hyperplane to segregate different classes. These extreme cases are called support vectors.

svm-2

If the number of input features is two, then the hyperplane is a line. If the number of input features is three, then the hyperplane becomes a 2-D plane.

Code

This model focuses on Cancer/Tumor type prediction. The result to be predicted is either Benign or Malignant.

The dataset has the following features:

'Clump', 'UnifSize','UnifShape','MargAdh', 'SingEpiSize', 'BareNuc', 'BlandChrom', 'NormNucl', 'Mit' and 'Class' is the target column.

There is one row named ID which contains the ID number of the patients but it is not used to train the model.

About

SVM Model for predicting if the tumor is Malignant or Benign

License:MIT License


Languages

Language:Jupyter Notebook 100.0%