Wildfires comprise one of the most destructive and lethal natural hazards, where various physical factors, such as low soil moisture, and human factors, such as the absence of preventive measures during wildfire season, can contribute to the generation of megafires. Although human activities such as arson make up the most frequent factor for a fire event, there are several natural hazards, such as lightning, heatwaves as an outcome of extreme temperature, and drought, that can also trigger or increase the probability of wildfires. On the other hand, subsequent secondary hazards such as floods and landslides can be effectuated in the long term as a result of wildfires. It is worth mentioning that the frequency of events of the aforesaid natural hazards has been increased intensively since 1980, according to the United Nations Office for Disaster Risk Reduction (UNDRR). The prediction of natural hazards is of outstanding importance, with the aim of disaster mitigation; thus, the utilization of Artificial Intelligence and Data Science is necessary. Various Machine Learning (ML) algorithms have been used in the literature to forecast wildfires and the aforementioned disaster chain, each one providing divergent predictions, in combination with Earth Observation (EO) indices, geospatial, and socio-economic datasets. The objective of this research is to expound the methods used in the literature to model and predict wildfires and their interconnected hazards by performing a review process.