Accuracy is a key factor in predicting a stock market. In the last decade, investors use the time series method is used to predict stock prices. But it needs improvement because the time series uses large data and time. In a given system, we are using Machine learning to predict stock prices. Machine learning is for enabling machines to learn like humans by collecting, storing, analyzing data, and developing a decision making on its own. Performing a search vector machine in a supervised machine learning algorithm can be done by studying an algorithm and by the statistical model. SVM use from classification as well as a regression problem. Ant it generally uses kernel trick for the transformation of data. It finds a moderate boundary between the possible outcomes. In statics linear regression is a linear approach to modeling the solar response and one or more explainer variables the process is called multiple linear regression for getting accurate output we implement machine learning along with surprised classifying this will be based on linear regression. The result will compare with real data and error will calculate. Linear regression techniques show an accuracy of 82%. Whereas, the proposed method shows an accuracy of 97% in prediction.