Supervised Machine Learning work on historical data e.g. Student learn under guidance of teacher in classroom. Supervised Learning algorithms are trained using labelled examples that is where we are aware output the input and we are aware about the output. The Learning algorithms receive input along with the corresponding output and the algorithms learn by comparing its actual output with the correct output to find the error.

Unsupervised Machine Learning is not an work on his data ,self learning & used itself. Unsupervised Machine Learning is used against the data that has not historical labels the system not told the right answer. The algorithms must figure out what is being shown the goal is to explore the data find some structure within the data.

Reinforcement Machine Learning is learn by error e.g. Google map. It is mostly used for the robotics, navigation and gaming with this learning algorithms discovers through the trail and error which action yield the greater reward.