If you don't know about Machine Learning (Maybe you from another planet) then the simple definition will be:
Machine Learning is a field of study that gives "Machines" (computers) the ability to learn without being explicitly programmed. — Arthur Samuel, 1959
There is one more very famous definition of Machine Learning which often confuses the newcomers:
A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E. — Tom Mitchell, 1997
I know it is confusing but let me break it down for you, and when we break this definition down your views on Machine Learning will be much clearer.
So Basically what the definition says is that there are three things on which Machine Learning Depends:
- Experience E - Data on which your program is going to learn
- Task T - Problem that your program is going to solve
- Performance Measure P - Formula that you are going to use to calculate performance. We can also consider it as accuracy.
So the definition explains that, if after learning from Experience E ( Data ) our problem will solve the Task T ( Problem ) with improved Performance Measure P ( Accuracy ) then we can say that our program is learning.
I think this will help you to understand the concept a little better. If still it, if not cleared just stick around you, will definitely get an Idea.
Creation of Machine Learning Application
Let's understand the flow of creation of Machine Learning application, observe the following diagram:
So understand the flow,
- We pass the Data to ML Algorithm
- ML Algorithm learn from the Data and creates a Solution
- We check for the Accuracy of the Solution
- If the Accuracy is Good enough then we launch the model for production
- Or if the Accuracy is Bad then we Analyze the Error and make some changes in the Algorithm.
Now, you can explain the definition of Machine Learning to others and you have also understood the basic process of the creation of a Machine Learning Application.
In the next blog, you are going to understand, " The different Types of Machine Learning ".