You should know: Introduction to Supervised machine learning
Before we start talking about supervised machine learning. Let us see what is Machine Learning (ML)!
A first definition of machine learning “is the field of study that gives computers the ability to learn without being explicitly programmed” by Arthur Samuel in 1959
Another more engineer oriented approach would be “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”. by Tom Mitchell in 1997
Tom Mitchell really made it sound hard but rhythmic. Let’s make it more simple! Let’s consider a checkers playing:
- The experience E: would be the experience of having the program play tens of thousands of games itself.
- The task T: would be the task of playing checkers,
- The performance measure P: will be the probability that wins the next game of checkers against some new opponent.
Machine Learning systems
Machine Learning systems can be classified depending on the amount and type of supervision they got during the training phase into Supervised machine learning, Unsupervised machine learning, and Reinforcement Learning. We can also add Semi-supervised learning which is the combination of Supervised and unsupervised machine learning.
Our focus in this article is Supervised machine learning!
What is supervised learning?
As you can see, supervised machine learning is a subcategory of machine learning, a type of artificial intelligence (AI) that allows software applications to more accurately predict outcomes without being explicitly programmed.
How does supervised learning work?
The first thing you need to know is that we provide the algorithm with pairs of inputs and desired outputs (target), As shown below.
That way when the algorithm is facing a new input it will find a way to produce the desired output.
Based on the type of our output (target) Supervised Machine Learning can be categorized into two.
Let’s take a closer look at Classification and Regression!
Regression
As we’ve seen before for supervised machine learning we provide pairs of Input and desired Output (Target) to our algorithm so that for a future new input he will find the way to produce the desired output. Here our desired output values are continuous values. A famous example in regression would be House pricing prediction.
Notice that the learning algorithm is trying to predict a “continuous (real) valued output”
Classification
As we’ve seen before for supervised machine learning we provide pairs of Input and desired Output (Target) to our algorithm so that for a future new input he will find the way to produce the desired output and for classification, our desired output values are discrete values.
Notice that the learning algorithm is trying to predict a “discrete (class) valued output (0 or 1) “
Classification is also categorized into types!
We note Binary classification, Multi-class classification, Multi-label classification
Some supervised Machine Learning Examples
Common Supervised machine learning algorithms:
❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁❁
That was a quick introduction to Supervised Machine Learning (❁´◡`❁)!
Hope you enjoyed reading it ╰(*°▽°*)╯!