According to Tom Mitchell, machine learning is:
“A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P if its performance at tasks in T, as measured by P, improves with experience E.”
In this definition:
Task T is what computer is seeking to reinforce. It can be whatever like prediction, classification, clustering, and so forth.
Experience E may also be coaching knowledge or input data through which the machine tries to gain knowledge of.
Performance P can be some aspect, like upgrades in accuracy or new skills that the laptop was beforehand blind to.
Machine learning itself involves two important add-ons: the learner and the reasoner.
Input/experience is given to the learner, who be trained some new knowledge.
Background knowledge may additionally accept to the learner for better finding out.
With the support of input and background, the knowledge learner generates the model.
The mannequin information about what's been learned from the enter and experience.
Now, the problem/task (i.E. Prediction, classification) is given to the reasoner.
With the help of trained model, the reasoner tries to generate the answer.
The solution/answer can be improved by using adding extra enter/expertise.
And so the cycle continues.