Sea-level rise affects the lives of people in coastal areas and it is causing the sea to inundate villages during high tide season which has changed people's lives because houses are destroyed and areas, where families gather, are being washed away. By predicting sea-level we can save their lives. Initially, for the local prediction, I have considered the following features which are affecting the most to rise of sea level: Local precipitation, Local temperature, Local population, Local CO2 concentration and for the global prediction, I have applied a different set of features: Global precipitation, Global temperature index, Global population, Global CO2 concentration. If we increased features than prediction will get even worse. So we have to choose features that affect the most to the prediction. The performance of linear regression algorithms under various conditions is derived using cross-validation on the latest 20% of the dataset, which was left out from the data-set. Other machine learning concept like PCA, GMM and SVM are also applied to understand their behavior.