Novel Based Ensemble Machine Learning Classifiers for Detecting Breast Cancer

Abstract

Nowadays, for many industries, innovation revolves around two technological improvements, Artificial Intelligence (AI) and machine learning (ML). ML, a subset of AI, is the science of designing and applying algorithms that can learn and work on any activity from past experiences. Of all the innovations in the field of ML models, the most significant ones have turned out to be in medicine and healthcare, since it has assisted doctors in the treatment of different types of diseases. Among them, early detection of breast cancer using ML algorithms has piqued the interest of researchers in this area. Hence, in this work, 20 ML classifiers are discussed and implemented in Wisconsin’s Breast Cancer dataset to classify breast cancer as malignant or benign. Out of 20, 9 algorithms are coded using Python in Colab notebooks and the remaining are executed using the Waikato Environment for Knowledge Analysis (WEKA) software. Among all, the stochastic gradient descent algorithm was found to yield the highest accuracy of 98%. The algorithms that gave the best results have been considered in the development of a novel ensemble model and the same was implemented in both WEKA and Python. The performance of the ensemble model in both platforms is compared based on metrics like accuracy, precision, recall, and sensitivity and investigated in detail. From this experimental comparative study, it was found that the ensemble model developed using Python has yielded an accuracy of 98.5% and that developed in the WEKA has yielded 97% accuracy.

SELF: a stacked-based ensemble learning framework for breast cancer classification

“…We summarize some of the recent machine learning, deep learning, ensemble learning techniques based on the breast cancer classification in Table 1. [2-4, 6, 9, 12, 14, 18, 20, 23, 24, 27-33, 36, 40, 47, 48] or textual datasets [1,5,10,17,21,34,41,42], and the researchers are working on them [1-6, 9, 12, 14, 17, 18, 20, 21, 23, 24, 27-34, 36, 40-42, 47, 48]. The existing works on classification are using the datasets with fewer number of sample images [1, 3, 5, 6, 17, 18, 21, 24, 28-30, 33, 34, 36, 41, 42, 47] that may not sufficient to train deep learning algorithms because training process of a deep learning network require a large amount of image data.…”

“…The existing works on classification are using the datasets with fewer number of sample images [1, 3, 5, 6, 17, 18, 21, 24, 28-30, 33, 34, 36, 41, 42, 47] that may not sufficient to train deep learning algorithms because training process of a deep learning network require a large amount of image data. The most frequently used datasets are the BreakHis [2,4,6,9,12,14,23,27,31,32,40,48] and the WBCD [1,5,10,17,21,34,41,42]. However, the WBCD dataset consists of only 569 or 699 instances with 32 features, while, the BreakHis dataset consists of 7909 images.…”

Machine Learning Algorithms for Breast Cancer Diagnosis: Challenges, Prospects and Future Research Directions

“…Machine learning is the science of designing and applying algorithms that can learn and work on any activity from past experience [13,29] . These algorithms have been applied in diagnosis of several cancer diseases such as leukemia, prostate, cervical and breast cancer [30] .…”

“…The causative agent of breast cancer is still under research, although some risk factors associated with this disease are well known and include age, gene, obesity, taking birth control pills and smoking [10] . The malignant tumor may begin in the cells of the breast, which then spreads to the surrounding tissues [11][12][13] . As explained by Rasool et al [14] , normal cells in the breast and other parts of the body grow and divide to form new cells as they are needed.…”

“…As discussed by Srinivas et al [13] , cancer can be classified into normal, benign and malignant tumors. It is also grouped into five stages (0-IV) depending on the size of the tumor and these stages can be identified as invasive and non-invasive.…”