Background: Kyasanur Forest Disease (KFD) is a haemorrhagic viral fever endemic to the forest regions of Karnataka. Many Machine Learning models have been designed and tested for the accurate prediction of the onset of disease. The dataset must be large enough to accurately predict the disease which impacts the early prevention and aid in the treatment strategies. However, the dataset of KFD is scarcely available and Machine Learning models fail to predict accurately. Few Shot Learning(FSL) is a unique kind of Machine Learning in which the model is trained to accomplish a task with a limited amount of training data. Objective: The key idea of this research work is to evaluate the efficiency and accuracy of various FSL methods in the prediction of the onset of KFD using the minimal dataset. Methodology: Many FSL methods from various FSL models have been trained and tested using the KFD dataset to compare the efficiency. Then the meta-accuracy and meta-loss of these methods are evaluated. Results: Among all, two methods Variational Encoders and Siamese networks show high meta-accuracy, 96% and 95% respectively. And their meta loss value is 0.08 which is very minimum. Compared with other methods, Variational Encoders and Siamese networks have shown better performance in predictive accuracy.