Published Paper Details:

Temperature prediction is an attempt to forecast the temperature of a particular region at a particular time. Temperature prediction is very important as it helps to plan various activities. A prime example of its application is in the field of agriculture. Knowing the range of temperature on a particular day can help a farmer prepare for the next day in advance and plan his activities to obtain optimal output. Temperature forecasts are required by utility companies because they help in estimating demand [1]. An example would be in helping people decide the clothes they should wear on a particular day and hence predict the upcoming demand. If a person expects a sudden drop in temperature, he/she can dress accordingly. On top of that, it helps in various aerial activities such as planning flights and optimizing air traffic. It can be applied to several other areas of interest including marine and military applications. Hence it becomes an essential part of our daily lives. This paper uses four different Machine Learning algorithms - Linear regression, KNN, RandomForest, Neural network - to predict temperatures at a particular time. Proposed algorithms are evaluated using standard evaluation metrics. The results of all the algorithms are analysed and the paper also discusses the performance of the proposed algorithm. The goal of this paper is to accurately predict the temperature in a particular region at a particular time, given the temperature history of that region over the past few days.