BREAST CANCER DETECTION WITH MACHINE LEARNING

Abstract

Accordin to the Breast Cancer Institute (BCI), breast cancer is
one of the most dangerous forms of cancers that, if diagnosed
and treated early enough, may be successfully treated for women
all over the world. It is believed by medical specialists that
detecting this cancer in its early stages can help save people's
lives by preventing it from spreading. This website, which covers
more than 120 distinct types of cancer and the genetic disord ers
that are connected with them, provides personalised therapy
suggestions based on the individual's medical history. Machine
learning algorithms are used to detect the vast majority of breast
cancers, which accounts for the majority of cases. This paper
presents an adaptive ensemble voting approach for newly
diagnosed breast cancer that was developed using the Wisconsin
Breast Cancer database and is based on a randomised controlled
experiment that was conducted using the Wisconsin Breast
Cancer database. The Wisconsin Breast Cancer database was
used in the research for this paper. The goal of this research is to
compare and explain how the ANN and logistic algorithms, when
used in conjunction with ensemble machine learning algorithms
for diagnosing breast cancer, generate greater outcomes when
the number of variables is reduced. The Wisconsin Diagnosis
Breast Cancer dataset, which was produced specifically fo r this
study, was used in this investigation. For the sake of comparison,
this study is being compared to other comparable studies that
have previously been published. When ANN metho dolo gy and
the logistic algorithm are coupled, they provide a classi fica t ion
accuracy rate of 98.50 percent when compared to another
machine learning technique, as demonstrated by a comparison to
another machine learning strategy (Figure 1).