Predicting flood hazard risk is crucial for reducing potential damage to infrastructures. This study uses machine learning algorithms to improve the accuracy of flood hazard risk predictions. We compare the performance of Long Short-Term Memory (LSTM) networks, Random Forest (RF), and Support Vector Machine (SVM) algorithms using a dataset to assess their effectiveness in predicting flood risk. The findings reveal that RF along with LSTM are the most accurate methods. These findings highlight the potential of machine learning algorithms, particularly RF and LSTM, in enhancing flood hazard risk analysis which offers valuable insights for risk mitigation strategies and infrastructure planning.