Global financial markets have dynamic nature that requires highly sophisticated strategies to predict and categorize investment strategies that can maximize returns, minimizing risks. The current study examines how machine learning algorithms can be used as a method of predicting and classifying investment strategies in international stock markets. As part of the training the study uses a detailed dataset with information on historical market prices, macroeconomic variables and technical analysis indicators. The effectiveness of key algorithms (Decision Trees, Random Forest) to predict the movement of asset prices and the classification of investment strategies (examples: growth, value, momentum and defensive investment strategies) is tested. The evaluation of the performance of these models would be based on accuracy, precision, recall and F1-score and the results are indicative of the possible application of machine learning in the decision-making process by investors. With this study, more sources on application of machine learning in the area of finance are created, and the results can be valuable to investors and a financial expert that wants to utilize tools based on AI technologies to make an optimal choice of their strategy considering fluctuation of the world market situation