BREAST CANCER DATA FEATURE SELECTION USING ENSEMBLE LIGHT GRADIENT BOOSTING TECHNIQUE

Abstract

Breast cancer is one of the highly dangerous diseases among the females around the world. The efficient and correct detection
of BC is big medical issue and many researchers proposed different diagnostic methods for detection of this disease, however
these existing methods still needed further improvement to correct and efficient detection of this disease. In this paper,
proposed a Ensemble Light Gradient Boosting Technique (LightGBM) new BC identification method by using machine
learning algorithm and clinical data. Feature Selection Algorithm is one more important step in the machine learning
classification process, as most of the time, there are many features in the dataset which are irrelevant or have the least
correlation with the output classes for example serial or ID number in any dataset. Such features affect the performance of the
machine learning classifiers. LightGBM Feature selection improves classification accuracy and reduces model computation
time.