Analysis of Lung Cancer Detection System for Better Identification Rate using Novel Support Vector Machine Algorithm in Comparison with Logistic Regression Algorithm

Authors

  • Y Kotesh kumar
  • R.Priyanka

Keywords:

Novel Support Vector Machine (SVM), Logistic Regression Algorithm, Lung Cancer, Image Processing, False detection, Machine Learning, Medical application

Abstract

Aim: This study's objective is to determine whether the novel Support Vector Machine (SVM) method can lower the false detection rate of lung cancer when compared to the logistic regression approach. Materials and Methods: Three lung cancer datasets accessible from kaggle and yielded a total of 304 samples. The SVM method is represented as group 1, while the logistic regression algorithm is represented by group 2. The G power calculation was done with an alpha of 0.05 and a power of 80%. Results: The accuracy of the SVM algorithm is 93.59% better than the 80.11% for the logistic regression approach. The SVM and logistics regression algorithm has the significance of (p=.0534). Conclusion: The SVM algorithm is proven to be much more accurate in this work when compared to the logistic regression algorithm.

Downloads

Published

2022-12-14