The Use of Data Mining in Predicting Breast Cancer A Case Study at Al-Durra Hospital – Khartoum

Abstract

The research presents an applied study on data mining and discovering knowledge from big data, which knowledge is often hidden in the midst of a huge amount of data.disease and provide results that contribute to reducing the spread of breast cancer in the coming years and the use of modern data mining techniques that facilitate data analysis, the study is interested in using techniques that extract and discover useful and independent knowledge through a large group of data, where we find that the emergence of new technologies. This led to drawing attention to the use of techniques in data mining in discovering solutions to some diseases, especially cancer diseases. First, I have to prepare the data obtained from the interview with the computer center of Al-Thaz Hospital Then, the application of data techniques that were selected in a specific sequence by reference to a group of reasons that can be summarized in two axes: the appropriateness of the method to the nature of the data and its compatibility with the research objectives in addition to the efficiency of pattern discovery, which led to the use of classification algorithms, namely TreeNaïve Bayes, neural networks, Decision and The aggregation algorithm, which is K-means, to achieve the objectives of the research when the data were selected, where we find that the most vulnerable age groups are among the 37-46 and most of them are women. The number of injuries was studied and distributed to the different states of Sudan. We find that the most affected states are the state of Khartoum, according to the type, where Females are the most infected with the disease, with 6942 infections, and men with the least, with 557 infections, and the years are more prevalent for the disease, as the year 2021 is the most prevalent, with 3736 infections, and the least year is 2012, with 138 infections.