Supervised Learning

The word supervised means observe and instruct the execution of a project task or an activity. But we are not supervising a person, yet we will be supervising a machine learning model. What supervised learning does is teaching the model and then model will predict the future instances using the knowledge that learnt. So, in general words model will be trained on a labeled dataset so it has enough train on each label. As example we can see below table.

Input1	Input2	Input3	Lable
10	1	2.1	Red
12	1	2.5	Red
11	2	3.1	Green
7	0	2.3	Yellow

Here according to Input 1,2, and 3 Label will be decided as Red, Green or Yellow. This is just the part of a table and it should have enough data for training. Once the model is trained it will be able to predict the label for given Input 1,2 and 3.

Also, in machine learning we use special terms to describe the dataset.

 

Attributes – Input1/2/3

Input1

Input2

Input3

 

Feature – Label

Lable

 

Observation – One single row

12

1

2.5

Red

 

Numerical value –Values with numbers

7

 

Categorical Value –Values with string label

Yellow

Normally this kind of data set will be stored in a .csv file or as comma separated values.

There are two types of supervised learning as Classification and Regression. We will talk about them in more details in next Articles.