Explainable artificial intelligence with scaling techniques to classify breast cancer images
Explainable Machine Learning for Multimedia Based Healthcare Applications, Springer International Publishing Ag, ss.99-137, 2023
- Yayın Türü: Kitapta Bölüm / Araştırma Kitabı
- Basım Tarihi: 2023
- Doi Numarası: 10.1007/978-3-031-38036-5_6
- Yayınevi: Springer International Publishing Ag
- Sayfa Sayıları: ss.99-137
- Anahtar Kelimeler: Breast cancer, Classification, Machine learning, Python language, Scaling techniques
- Recep Tayyip Erdoğan Üniversitesi Adresli: Evet
Özet
According to the Breast Cancer Institute, one of the most hazardous diseases for women is breast cancer (BCI). Clinical specialists claim that detecting this malignancy early can save lives. More than 120 distinct tumors and related hereditary illnesses have individualized resources available, according to cancer.net. To diagnose breast cancer, machine learning techniques are mostly used. This research projects the use of eight machine learning (ML) approaches to accurately detect and classify breast cancer images as benign and malignant. Using Wisconsin Diagnostic Breast Cancer (WDBC) Dataset, various scaling techniques were used on the eight different algorithms to build the projected model. Various libraries in Python programing language were used in Jupyter environments to achieve classification objectives. The results obtained from this implementation show that the Random Forest (RF) model gave the best classification results with 98.60% accuracy, 98.21% f1-score, 98.21% precision, 98.21% sensitivity, and false positive rate (FPR) of 1 on L2 scaling from other ML approaches implemented.