Wildfires are catastrophic phenomena that occur quite frequently on the west coast of the United States. In a 2003 San Diego study, the total economic costs of wildfires were estimated to be $2.45 billion, with around 376,000 acres burned. Due to significant loss of life, property, and environmental damage occurring every year, it is crucial that this problem is approached early, and predicted beforehand. Wildfires occur when all necessary physical conditions such as temperature, wind, humidity, and relative humidity come together in a susceptible area. This combination of conditions coupled with a source of ignition and adequate supply of oxygen is sufficient to cause the fire to spread. This project focuses on understanding the effects of the physical and environmental conditions such as temperature, wind, humidity and relative humidity on predicting the occurrence of fire in a given area. The data for this project is taken from the Portugal Fire Dataset. The four main modeling methods that were used for analysis are Linear Regression, Decision Trees, Support Vector Machines, and Artificial Neural Networks. These models were then tested on a newly self-created dataset compiled from fires primarily on the west coast of the United States since 2000. The predicted results were then compared with the actual occurrence of forest fires for validation. It was found that the Support Vector Machine model with an accuracy of prediction of 76% performed the best. This project opens multiple possibilities for effective evacuation methods when risk of a forest fire is high.