What is Machine Learning?
It is the science of getting computers to act without being explicitly programmed. In simple words, Machine Learning constructs the models and algorithms which will not only predict on data but also learn from it.
It involves but not limited to this:
a) trying to detect an output based on some relationship between input and output.
b) trying to identify some structures in the data and then predicting.
Real Life Examples
Here are the few examples where there is machine learning behind the scenes but probably you never got to know!
a) Facebook Image Detection Feature:- When you upload an image on facebook, it detects the objects by pointing out. This is nothing other than machine learning usage.
b) Email Spam Classification:- Distinguish an email from spam vs non spam is done using machine learning. This is done by classifying the words in a list as spam against others being non spam. Some famous algorithms for this are Support Vector Machine Algorithm (SVM), Bayesian Algorithm etc.
As I was mentioning, ML is the act of getting computers to act but how will the computers act? A computer program is said to learn from
- Experience E
- with respect to some class of tasks T &
- performance measure P
“if performance at tasks in T -> as measured by P -> improves with experience E”
Example: E: experience of playing many games of checkers.
T: tasks of playing checkers.
P: probability that program will win next game.
Machine Learning Classification
Machine Learning can be broadly classified into two categories Supervised and Unsupervised Learning.
In supervised learning: There is an input set of data and the aim is to predict the output as we have an idea about the relationship between the input and output.
In un-supervised learning: Aim is to find some structure in data. It allows us to approach problems with little or no idea what our results should look like.