Comparative Study of Anemia Classification Algorithms for International and Newly CBC Datasets

Data generated from modern applications and the internet in healthcare is extensive rapidly expanding. Therefore, one of significant success factors for any application understanding extracting meaningful information using digital analytics tools. These tools will positively impact application's performance handle challenges that can be faced to create highly consistent, logical, information-rich summaries. This paper contains three main objectives: First, it provides several methodologies help analyze datasets extract useful them as preprocessing steps classification model determine dataset characteristics. Also, this a comparative study algorithms by testing 12 different classifiers two international provide an accurate indicator their efficiency future possibility combining efficient achieve better results. Finally, building CBC first time Iraq helps detect blood diseases hospitals. The outcome analysis step used researchers select best system structure according characteristics each more organized thorough test results, four achieved accuracy (Logitboost, Random Forest, XGBoost, Multilayer Perceptron). Then use Logitboost algorithm classify these new datasets. In addition, directions, investigate utilize benefits overcome disadvantages.