Supervised and Unsupervised Machine Learning

Introduction
American pioneer Arthur Samuel created the phrase "machine learning" in 1959. stated
that machine learning is "the field of study that gives computers the ability to learn
without being explicitly programmed"
When you tag a face of Facebook photo, it is machine that is running behind the scenes
and identifying faces in the picture. There are many programmes that can identify things
like animals, dogs, bottles, autos, etc. Our roadways are currently being used by
autonomous vehicles that can steer by detecting objects in real time. When you're on the
road, you use Google Directions to research the current traffic condition and take the
best route that Google suggests at the time. This is yet another real-time application of
object detecting technology. There are several applications of machine learning that we
use practically today. In fact, each one of us use machine learning in many parts of our
lives, even without our knowledge.
Categories of Machine Learning
Machine learning is broadly categorised under the following headings:
 Supervised learning
 Unsupervised learning
 Reinforcement learning
 Deep learning
 Deep reinforcement learning
Machine learning evolved from supervised learning to deep reinforcement learning
Supervised Machine Learning
Supervised learning is like training a child to walk. You will hold the child’s hand, show
him how to take his foot forward, walk yourself for demonstration and so on, until the
child learns to walk on his own. The two common type of supervised learning are
regression (where the outputs are real-value) and classification (where the output are
labels, as in spam filtering)
Similarly, in the case of supervised learning, you give concrete known examples to the
computer. You say that for given feature value P1 the output is Q1, for P2 it is Q2, for P3
it is Q3, and so on. Based on this data, you let the computer figure out relationship
between P and Q. Once the machine trained in this way with sufficient number of data
points, now you would ask machine to predict B for a given A. Assuming that you know
the real value of Q for given P, you will be able to know whether the machine’s
prediction is correct.
Thus, you will test whether the machine has learned by using the known test data. Once
you are satisfied that the machine is able to do the predictions with a desired level of
accuracy you can stop further training of machine. Now, you can safely use the machine
to do the prediction on unknown data points, or ask the machine to predict Q for given P
for which you do not know the real value of Q.
Objects of a similar kind can be grouped together using machine learning techniques.
For instance, you might want to divide the 100 students into three groups based on their
heights: short, medium, and long. You will assign each kid to the appropriate category
after measuring their height. Now, you will measure a new student's height and place
him in the right category. By adhering to the regression training principles, you can
teach the computer to categorise students according to their height, one of their
features. The machine will be able to appropriately categorise any unknown new
student after it learns how the groups are created. Once again, you would use the test
data to verify that the machine has learned your technique of classification before
putting the developed model in production.
Unsupervised Machine Learning
Unsupervised learning is when we are given a collection of unlabelled data. Which we
wish to analyse and discover pattern within. The two most important example are
dimension reduction and clustering.
In unsupervised learning, we do not specify a target variable to the machine, rather we
ask machine What can you tell me about A? More specifically, we may ask questions
such as given a huge data set A, what are the five best groups we can make out of A? or
what features occur together most frequently in A? To arrive at the answer to such
questions, you can understand that the number of data points that the machine would
require to know a strategy would be very large. In case of supervised learning, the
machine can be trained with even about thousands of data points that is reasonably
accepted for learning starts in a few millions. These days, the data is generally
abundantly available. The data ideally requires arrangement. However, the amount of
data that is continuously flowing in a social area network, in most cases data
arrangement is an impossible task.
“Unsupervised learning is a type of machine learning in which models are trained using
unlabelled dataset and are allowed to act on that data without any supervision.”
Unsupervised learning cannot be directly applied to a regression or classification
problem because unlike supervised learning, we have the input data but no
corresponding output data. The goal of unsupervised learning is to find the underlying
structure of dataset, group that data according to similarities and represent that data in
compressed format.
Consider the following scenario: An input dataset including images of various breeds of
cats and dogs is provided to the unsupervised learning algorithm. The algorithm is
never trained on the provided dataset, thus it has no knowledge of its characteristics.
The unsupervised learning algorithm will carry out this operation by grouping the
image collection into groups based on visual similarities.
Following are some main reasons which describe the importance of unsupervised
learning
 Unsupervised learning is helpful for finding useful insights from the data
 Unsupervised learning is much similar as a human learns to think by their own
experiences, which makes it closer to the real AI
 Unsupervised learning works on unlabelled and uncategorized data which make
unsupervised learning more important
 In real-world, we do not always have input data with the corresponding output
so to solve such cases, we need unsupervised learning.
Type of Unsupervised learning algorithm
Clustering
Clustering is a method of grouping the object into clusters such that objects with most
similarities remains into a group and has less or no similarities with objects of another
group. Cluster analysis find the commonalities between the data objects and categorises
them as per the presence and absence of those commonalities
Association
An association rule is an unsupervised learning method which is used for finding the
relationships between variables in the large database. It determines the set of items that
occurs together in the dataset. Association rule is makes marketing strategy more
effective. Such as people who buy P item are also tend to purchase Q item. A typical
example of association rule is market basket analysis.
Below is the list of some popular unsupervised learning algorithm:
 K-means clustering
 KNN (k-nearest neighbours)
 Hierarchal clustering
 Anomaly detection
 Principal component analysis
 Independent component analysis