The Global Land Surface Temperature is the radiative skin temperature of ground, depending on factors, which includes the albedo, the vegetation covers and the soil moisture. To predict the changes in temperature in a particular region is becoming increasingly important to capture the future trends in that region. Machine Learning is a specialized branch of Artificial Intelligence (AI), which gives computers the power to learn and make predictions from the data, without being explicitly programmed. In this work, ,machine learning Approach such as Linear Regression, K-Nearest Neighbor Decision tree and Ada boost algorithms for Global mean Temperatures is proposed. This approach will help to predict the temperature, which is of great requirement as the problem of global warming is increasing day by day. Temperatures are collected from different cities and prediction is done using this approach. The proposed ML approach is based on the models which provide good performance in terms of model evaluation parameters like Accuracy. Experimental result show the better performance of our system.