The purpose of this paper is to analyze and examine the effect of machine learning algorithms alternatives on the prediction accuracy of going concern opinion and which one is more effective in predicting the accuracy of going concern opinion. To achieve this purpose, the research will address the accuracy of going concern opinion from a professional view, determinants of the accuracy of going concern opinion, measurements of accuracy of going concern opinion, machine learning from a professional view, and analysis the effect of the machine learning algorithms on the accuracy of going concern opinion.
 
In order to test the research hypotheses, the researcher will use the decision trees (DT), logistic regression, support vector machines (SVM). The sample used in the current study consists of 87 non-financial companies listed in Egyptian Stock Exchange during the period (2019-2021). The research concludes that SVM and Logistic regression has the highest accuracy to predict going concern doubts, where the accuracy rate is 86%, then the decision tree model doubts, where the accuracy rate is 79 %.
 
In light of the research objectives and its problem, and the results it concluded, the research recommends that auditors should be interested in developing their skills to be able to use artificial intelligence, such as machine learning, in issuing audit opinion, as they face some difficulties in using artificial intelligence in the audit field.
Regarding the proposed research areas, the most important of them are the following: (a) the effect of using data analytics on the prediction accuracy of going concern opinion, (b) The effect of artificial intelligence technologies on audit evidence, (c) The effect of machine learning on detecting misstatements on financial statements.