Comparative Analysis of Accuracy, Sensitivity & Specificity of Novel Decision Tree Algorithm to Maximise the Early Detection Rate of Lung Cancer in Comparison with Logistic Regression Algorithm

Abstract

Aim: The objective of this research work is to maximise the early detection rate of lung cancer using the novel decision tree algorithm in comparison with the logistic regression algorithm. Methods and Materials: A total of 304 samples are collected from three lung cancer datasets available in kaggle. Group 1 represents the novel decision rate algorithm and group 2 represents the logistic regression algorithm.The G power calculation was done with 80 % of power and alpha of 0.05. Results: Novel decision tree has achieved the significance accuracy of 94.86 % compared to 80.11 %, by logistic regression algorithm. The novel decision tree algorithm has achieved the significance value of 0.044 when compared to the logistic regression algorithm. Conclusion: In this research, it was observed that the decision tree method outperforms the logistic regression approach in detecting lung cancer in the datasets that were considered.