Recent advancement in information technology have improved the design and development of disease prediction systems significantly. The disease prediction system is useful in diagnosing diseases by analyzing medical data. In this digital world, disease prediction systems are extremely important, especially during pandemic situations when physicians are in high demand and people are unable to reach hospitals to monitor and diagnose their health conditions. Many medical expert systems and disease prediction systems have been published in recent years. Still, there is a gap for people to have an effective disease prediction system to predict a patient's disease and severity level at the right time. Predicting the impact level of disease in the human body is considered one of the most difficult issues nowadays due to the increase in voluminous medical data with various new symptoms. This paper presents a survey of machine learning and deep learning techniques for dengue fever prediction. It covers methods like association rule mining, decision trees, clustering, and neural networks, assessing their accuracy and practicality. Challenges like data imbalance and environmental factors are discussed, offering insights for future research in disease prediction.