Machine learning

Introduction:

Machine learning is a
many-faceted phenomenon. Learning process include the acquirement of new information
and knowledge, the development and improvement of skills through continuous work
and the discovery of new valuable knowledge through thinking and
experimentation. Artificial intelligence is the field of computer science that
focuses on the creation of intelligent machines that work and behaves like
humans. Machines can react like humans only if they have handsome knowledge
about the world.

Some activities of computers
with artificial intelligence are designed to include speech recognition,
learning, planning, problem solving. Machine learning is an application of
artificial intelligence that provides systems the ability to automatically
learn and improve from experience without being explicitly programmed. Machine learning focuses on the development of
computer programs that can access data and use it learn
for themselves.

The process of learning begins with
observations or data, such as examples, direct experience, or instruction, in
order to look for patterns in data and make better decisions in the future
based on the examples that we provide. The primary aim is to allow the computers learn automatically without
human intervention or assistance and adjust actions accordingly.

Objective:

Machine
learning focus on three primary areas of research:

·        
Task oriented
studies:

Task oriented
studies focus on the development and analysis of the learning system which are
helpful in improving the performance of set of task. It is also known as
engineering approach

·        
Cognitive
Simulation:

The analysis
and act of human learning process is known as cognitive simulation.

·        
Theoretical
Analysis:

The learning
processes which are independent of domains.

 

There are multiple applications for
Machine Learning (ML), the most significant of which is data mining. People are
often prone to making mistakes during analyses or, possibly, when trying to
establish relationships between multiple features. This makes it difficult for
them to find solutions to certain problems. Machine learning can often be
successfully applied to these problems, improving the efficiency of systems and
the designs of machines.

#Re phrase it….

Another kind of machine learning is
reinforcement learning (Barto & Sutton, 1997). The training information
provided to the learning system by the environment (external trainer) is in the
form of a scalar reinforcement signal that constitutes a measure of how well
the system operates. The learner is not told which actions to take, but rather
must discover which actions yield the best reward, by trying each action in
turn. Numerous ML applications involve tasks that can be set up as supervised.
In the present paper, we have concentrated on the techniques necessary to do
this.

 

Our
first section covers the introduction of Machine Learning, its applications and
core objective then wide-ranging issues of supervised and un-supervised machine
learning such as data pre-processing and feature selection are described. Related
work is described in section 3, whereas methodology is described in section 4. Results
and discussion is described in section 5. Finally, the last section concludes
this work

Related Work:

The
field of Machine Learning relies upon nature’s bounty for both inspiration and
mechanism. Many machine learning systems are borrowed from current thinking in
cognitive science and unlighted interest in neural networks. Another area where
example has been tapped is in work on genetic algorithm and genetics based
learning.

Machine Learning approach to genetic algorithms

This
special double issue of Machine Learning is devoted to papers concerning
genetic algorithms and genetics-based learning systems. Simply stated, genetic
algorithms are probabilistic search procedures designed to work on large spaces
involving states that can be represented by strings. These methods are
inherently parallel, using a distributed set of samples from the space (a
population of strings) to generate a new set of samples. They also exhibit a
more subtle implicit parallelism.

Although
there are a number of different types of genetics-based machine learning
systems, in this issue we concentrate on classifier systems and their
derivatives. Classifier systems are parallel production systems that have been
designed to exploit the implicit parallelism of genetic algorithms. All
interactions are via standardized messages, so that conditions are simply
defined in terms of the messages they accept and actions are defined in terms
of the messages they send. The resulting systems are computationally complete,
and the simple syntax makes it easy for a genetic algorithm to discover building
blocks appropriate for the construction of new candidate rules 1,2

Machine Learning approach to text categorization

Machine
Learning is widely used in the automated text categorization. In the last ten
years content-based document management tasks have gained a prominent status in
the information systems ?eld, due to the increased availability of documents in
digital form and the ensuing need to access them in ?exible ways.

Text
categorization is the task of assigning a Boolean value to each pair hdj,cii ? D×C, where D is a
domain of documents and C = {c1,…,c|C|} is a set of prede?ned categories. A
value of T assigned to hdj,cii indicates a decision to ?le dj under ci, while a
value of F indicates a decision not to ?le dj under ci. More formally, the task
is to approximate the unknown target function ?? : D ×C ? {T,F} (that describes
how documents ought to be classi?ed) by means of a function ? : D × C ? {T,F}
called the classi?er (aka rule, or hypothesis, or model) such that ?? and ?
“coincide as much as possible”. The categories are just symbolic labels, and no
additional knowledge of their meaning is available.

In
the ’80s the most popular approach (at least in operational settings) for the
creation of automatic document classi?ers consisted in manually building, by
means of knowledge engineering (KE) techniques, an expert system capable of
taking TC decisions. Such an expert system would typically consist of a set of
manually de?ned logical rules, one per category, of type

if
hDNF formulai then hcategoryi

A DNF
(“disjunctive normal form”) formula is a disjunction of conjunctive clauses;
the document is classi?ed under hcategoryi i? it satis?es the formula, i.e. i?
it satis?es at least one of the clauses. it was originally suggested that this
approach can give very good e?ectiveness results: Hayes et al. 1990 3

Machine Learning approach to selection of relevant
information

Machine
Learning works for handling datasets containing large amount of irrelevant
information. Machine learning aims to address larger, more complex tasks, the
problem of focusing on the most relevant information in a potentially
overwhelming quantity of data has become increasingly important. For instance,
data mining of corporate or scientific records often involves dealing with both
many features and many examples, and the internet and World Wide Web have put a
huge volume of low-quality information at the easy access of a learning system.
Similar issues arise in the personalization of filtering systems for
information retrieval, electronic mail, netnews, and the like.5