In recent years, the use machine learning techniques in health diagnosis has received increased interest particularly in the detection of cancer at the early stage. In this research work, the emphasis is laid on the differences in performance of the developed and optimised MLP in comparison with other machine learning techniques for cancer diagnosis. This study’s main objective is to identify the optimal predictive model by establishing the accuracy, precision, recall, F1 score, and the AUC. In the context of the study, several sources of data are used to enhance the evaluation framework; more specifically, the Brain tumor Kaggle Dataset is used. The hyper-parameters in the MLP model were efficiently tuned for better predictive capabilities using the methods such as the Grid search method and cross-validation. Most popular Comparative algorithms used are Support Vector Machines (SVM), Random Forest, K- Nearest Neighbours (KNN), Decision Trees, Logistic Regression, Gradient Boosting Machines, Naive Bayes, and Gradient Boosting. The accuracy analysis presented that the optimized MLP yielded the best accuracy of 91% as compared to SVM at 88; the Random Forest at 89% and Gradient Boosting at 87%. Besides that, MLP showed the accuracy of 0. 92, compared to 0. For Decision Trees, it has reached a score of 88 while for committee-based Justification it is at score 0. 90 for Logistic Regression. The actualization for MLP was 0. 90 which was better to KNN which had 0. 85 and Naïve Bayes which also had 0. 87. The F1 score for MLP was 0.91, while Gradient Boosting Machines scored 0.88 and Random Forest 0.89. This is because early cancer detection is essential in the treatment of this deadly disease since the chances of survival are boosted greatly. It implies that detecting the disease at a time when the chances of the existence of an effective remedy is high hence helping in the reduction of the death rate and enhancing the well-being of the patients.
Yashdeep Raj1, Kanwal Garg2, Tajinder Kumar3
Kuruksetra University, India1,2, Jai Parkash Mukand Lal Innovative Engineering and Technology Institute, India3
Cancer Detection, Multi-Layer Perceptron, Machine Learning, Predictive Modelling, and Random Forest
| January | February | March | April | May | June | July | August | September | October | November | December |
| 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 0 | 0 |
| Published By :
ICTACT
Published In :
ICTACT Journal on Data Science and Machine Learning
( Volume: 5 , Issue: 4 , Pages: 688 - 696 )
Date of Publication :
September 2024
Page Views :
13
Full Text Views :
2
|