Linear Regression is a type of supervised learning model which is used for forecasting. In the Supervised Learning model, we use training data to build the model and then use test data to test its accuracy.
Linear Regression shows the relationships between a set of independent variables to that of the dependent variable.
Linear Regression is the plotting of a straight line of a form y=mx+c such that it predicts the data points. In other words, if our model is well trained using Linear Regression then, in that case, the predicted point will lie on the regression line.
Let's suppose that we have 2 axis x & y where the x-axis has independent variables and the y-axis has dependent variables and here our aim is to draw the Regression Line. If we have a data point on the x-axis which increases and is independent in nature, similarly we have a data point on the y-axis which is also increasing and is dependent in nature thus we will get a positive regression line.
Suppose that the data point of the y-axis is decreasing then we will get the negative regression line.