Prediction of weather is an attempt done by meteorologists to forecast the weather conditions of an area at some time in the future that may be expected. The parameters of the climatic condition are based on the humidity, wind, temperature, rainfall and size of data set. Snowfall has a tremendous effect on the livelihood and the socio-economic development of the native individuals. It extends the permanency of glaciers and thereby providing adequate water for drinking, irrigation and hydro-power generation. Furthermore, it enhances business enterprise and provides diversity advantages. Snowfall is a blessing however if it is not correctly predicted, it can become a curse. The cost of clearing the snow is high both in terms of cost and time. Predicting snowfall for a short-term is effective in scheduling when and how to clear the snow and also necessary to decrease the cost of clearing it. Since the weather is a non-linear phenomenon and its randomness is very difficult to comprehend both in terms of complexity and technology, deep-learning has been applied as it is proficient in understanding the non-linearity and also comparatively robust to perturbations. This paper proposes a deep learning algorithm (LONG SHORT-TERM MEMORY) which is employed on three different datasets obtained from three different locations of Jammu and Kashmir. The results found from this experiment depicted that LSTM was capable of forecasting the snowfall with substantial accuracy.