Cancer is the uncontrolled growth of cells that spread to other parts of the body. There are different types of cancer that are named after the organ they originate from. One of them is blood cancer or leukemia, which is bone marrow cancer caused by genetic mutations. According to data from Global Cancer Statistics in 2020, there were an estimated 19.3 million new cancer cases and 10 million cancer deaths, and it is estimated that by 2040 it will increase globally by 47% from 19.3 million to 28.4 million new cancer cases. Leukemia is one type of cancer with the ninth rank in Indonesia in 2020, there are 14,979 new cases and 11,530 cases of death caused by leukemia. One of the efforts to prevent leukemia can be done by diagnosing the acute leukemia category using DNA and genetic information. The purpose of this study is to analyze the comparative performance between Random Forest and Adaptive Boosting methods in predicting leukemia types using microarray datasets to determine which method is more effective in performing classification. In this study, the dataset used is gene expression in bone marrow and blood consisting of two categories of acute leukemia, namely Acute Myeloid Leukemia (AML) and Acute Lymphoblastic Leukemia (ALL) obtained with DNA microarray technology. These genes will be classified using Random Forest and Adaboost methods to predict acute leukemia categories. The results of the analysis process show that the random forest method is a better method for predicting acute leukemia with an Area Under Curve value of 100%, Accuracy 92.9%, Precision 93.7%, Recall 92.9%, and F1-Score 92.7% compared to the AdaBoost method with an Area Under Curve value of 83.3%, Accuracy 85.7%, Precision 88.6%, Recall 85.7%, and F1-Score 85.1%.