| Autor: |
Maheswari, B. Uma, Guhan, T., Britto, Christopher Francis, Sheeba, Adlin, Rajakumar, M. P., Pratyush, Kumar |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2023, Vol. 2831 Issue 1, p1-7, 7p |
| Abstrakt: |
Throughout the entire lifetime of women, it is observed that nearly 9% of women are detected with the diagnosis of Breast Cancer (BC) which is considered as a deadliest disease that maximally leads to death. Since, lung cancer is considered as the most dangerous diseases that ranks first which leads to death with which the breast cancer ranks second in the moderately developed and well-developed countries. The characterization of breast cancer is made by the gene mutation, continuous pain, size variation, darkening of skin (reddishness) and variation in the texture of breast skin. Classifying the breast cancer images leads the clinicians to detect the objective and systematic prognostic, commonly the most frequent categorization includes malignant and benign sort of cancer. Nowadays, the strategies of machine and deep learning are widely applied in categorizing and classifying the different types of breast cancer. These algorithms, lends the maximal rate of accuracy over classification and the efficient capability of diagnosis. In this proposed study, the disclosure of two various classifiers has been made that is inclusive of k-nearest neighbor (kNN) and Naive Bayes (NB) for classifying the breast cancer. The performance comparisons are also being made between these two proposed classifiers based on the performance metrics evaluation such as accuracy and validation extended by the k-fold cross validation technique. It is being inferred that the acquired results proves that the kNN produces maximal classification accuracy with 97.68% whereas the NB classifier produces 96.21% of classification accuracy. Since, both the classifiers have worked with minimal error rate. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
