A review of early breast cancer prediction techniques and evaluation of these techniques based on appropriate criteria.

Document Type : Original Article

Authors
Farshid Vazifehdoost 1
Somayeh Kadkhoda Dehkhani 2
Maedeh Rahmani 3
Mehdi Ghasemi 4
Hamid Zangiabadi zadeh 5
1 Graduated with a Masters degree in Computer Engineering, Artificial Intelligence and Robotics, and Information Technology, Electronic Business, from Payame Noor University.
2 Graduated with a Masters degree in Computer Engineering,Artificial Intelligence and Robotics
3 Masters degree student in computer engineering, software direction, Payame Noor University
4 Masters degree student in computer engineering at Payame Noor University
5 Masters degree student in Computer Engineering, Artificial Intelligence and Robotics, Payame Noor University, Kish International Center
Abstract
The incidence of breast cancer has been increasing steadily over the years due to changes in lifestyle and environment. Currently, breast cancer is one of the leading causes of cancer mortality among women, making it a critical concern for global public health. Therefore, the development of an automated breast cancer detection system is of great importance in the medical community. This article discusses the general concepts of breast cancer, data mining, and machine learning, and also introduces useful machine learning techniques that can be useful for breast cancer classification and prediction. In the main part of the article, the evaluation and comparison of these techniques were reviewed and performed based on appropriate criteria of accuracy, precision, recall, specificity, and F1 score. The results of the evaluation of these classifiers in research whose experiments were conducted on standard datasets showed that among the introduced techniques, support vector machine, random forest, and decision tree, respectively, have stronger prediction than the other classifiers.
Keywords
Machine learning
Classification
Data
Breast cancer
Artificial intelligence
Subjects