Machine Learning is basically learning patterns and making prediction.
There are basically two types of machine learning algorithms :-
- Supervised learning
- Unsupervised learning
In this type of machine learning algorithm we provide labels or target values. For example :- we give a picture of dog and also label it that it is dog.
In this type we do not provide labels or target values. For example :- given balls of different color then using clustering our algorithm can make different clusters of each color. While predicting only check to which cluster it belongs.
Preparing Data for Algorithms
We divide our data in two parts one training and other testing. If we have less data then overfitting is a common problem. To avoid it we use K fold validation. In this we divide our data in equal K parts. We keep one part for testing and rest for training. We do this for every part.