Machine learning

We are going to
toward the time zone of artificial intelligence where everything is controlled
and handled by the machine. Machine learning is the subset of artificial
intelligence where we teach the machine to learn by itself without the help any
external source. In machine learning we teach the machine to learn from its
previous data and try to improve its result in future by taking lesson from its
previous decision. Part of machine learning includes the uses of tools, methods
and techniques which help it form better results. These methods and algorithms
provide machine and us a new approach to explore the new knowledge from and
given data or the by exploiting the traditional datasets. In some situation, we
try to record the behaviour and then model that behaviour. In turn modelling
stimulate the people to have a better understanding of the situation. Machine
learning method have a slight have a history of statistics. Its helpful for
exploring more complicated learning model to take out the true message hidden
in large amount of data. 

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

Although both machine
learning technologies and traditional statistics tools can be applied in data
analysis, their fundamental principles and characteristics have a great
different. As compared to statistics data analysis, the exclusive advantages of
machine learning includes enumerated benefits which are : we can process big
data and real-time data streams with mixed values types, we can select from
different learning models and controlling parameters to capture the non-linear
or high-order structure in data, we can also recognize complicated patterns
that cannot be represented in different mathematical terms, visualization of
the data for making a prediction and we can also integrate the learning models
with other different databases management system.

Learning
is a continuous process but in machine also we try to stimulate this learning
process alike humans. For machine basically learning consist of 3 types which
are supervised, unsupervised and reinforcement learning. All these play a major
role to designing a learning models. Where First, in supervised
learning, the computer is provided with example inputs that are labeled with
their desired outputs. The purpose of this method is for the algorithm to be
able to “learn” by comparing its actual output with the “taught” outputs to
find errors, and modify the model accordingly. Supervised learning therefore
uses patterns to predict label values on additional unlabeled data. A common
use case of supervised learning is to use historical data to predict
statistically likely future events. Second, In unsupervised learning, data is
unlabeled, so the learning algorithm is left to find commonalities among its
input data. As unlabeled data are more abundant than labeled data, machine
learning methods that facilitate unsupervised learning are particularly
valuable. In Third, Reinforcement Learning – a goal-oriented learning based
on interaction with environment. Reinforcement Learning is said to be
the hope of true artificial intelligence. And it is rightly said so, because
the potential that Reinforcement Learning possesses is immense. It allows machines and software
agents to automatically determine the ideal behavior within a specific context,
in order to maximize its performance. Simple reward feedback is required for
the agent to learn its behavior; this is known as the reinforcement signal.