Sunteți pe pagina 1din 4

UNIT 6: Artificial Intelligence

Concepts of Artificial Intelligence

Artificial Intelligence - is branch of Science which deals with helping machines find
solutions to complex problems in a more human-like fashion.
- Borrowing characteristics from human intelligence.
- Then, applying them as algorithms in a computer-friendly way.

What is Artificial Intelligence?


It is a broad field, and many mean different things to different people.
It is concerned with getting computers to do tasks that require human intelligence.
It deals with the simulation of the intelligent human behavior to a machine or computer.
It is also concerned with the difficult tasks such as complex arithmetic, which seem to
require complex and sophisticated reasoning processes and knowledge
It is a field that overlaps with Computer Science rather than being a strict subfield.
Supports the field of Computer Science, Math, Psychology, Cognition, Biology and
Philosophy.

What are the possible reasons why people consider Artificial Intelligence?
To simply understand human intelligence.
To have smarter programs.
To be set out to solve domain specific problems, by using algorithms.
Techniques of Artificial Intelligence
a) Knowledge Representation - These techniques that have evolved are concerned with
how we represent, manipulate, and reason with knowledge in order to solve problems.
Knowledge must be meaningfully represented so that we know how it relates back to the
real world. A knowledge representation scheme provides a mapping from features of the
world to the formal language. The formal language will just capture certain aspects of
the world, which we believe are important to our problem. (Sometimes, we may miss out
crucial aspects and so fail to solve our problem. When we talk about the semantics of
representation languages situations like manipulating formal language using a computer
to make sure that we still have meaningful expressions that can be mapped back to the
real world.
b) Search - It is another crucial general technique required when writing AI program. Often
there is no direct way to find a solution to some problems. However, generating
possibilities is next in line. Developing good ways to search through these possibilities
for a good solution is therefore vital. And here are the two major ways to search:

The Brute Force Techniques (very strong or forceful) where you generate and try
out every possible solution may work. But this technique is often very inefficient,
as there are too many possibilities to try.
While the Heuristic Techniques (it means involving /serving as an aid to
problem solving by experimental and/or trial-and-error) are often better where
you only try the options which you think (based on your current best guess) are
most likely to lead to a good.
c) Fuzzy Logic - system of logic in which statements can be true, false or any continuum of
values in between. It uses rules but assumes that conditions and conclusions of rules are
stated in an approximate (almost correct or exact: close in amount but not precise) way
without forcing people to pretend that their perceptions and beliefs are unrealistically
precise. Decision-making answers are always clouded by may be as a pretty good
guesses. In this technique, everything has a degree of certainty. This degree of certainty
has a value from 0 to 1. 0 for FALSE and 1 for TRUE. **The numbers between 0 and 1
are used to indicate what extent an answer is uncertain. While the Heuristic Techniques
are often better where you only try the options which you think (based on your current
best guess) are most likely to lead to a good solution.
d) Intelligent Computer Aided Instruction (ICAI) - It refers to the work of machines that
can tutor humans.
Uses of ICAI
For Problem Solving
Simulation
Discovering
Drill And Practice
Games
e) Computer Vision or Vision System - It includes hardware and software that permits
computers to capture, store, and manipulate visual images and pictures. Visual
Recognition, It is the addition of some forms of computer intelligence and decisionmaking to digitalized visual information received from a machine sensor. Combined
Information, It is used to perform or control such operation as robotics movement,
conveyer speed, and production line quality control.
f) Machine Learning - It refers to a set of methods that attempt to teach computers solve a
problem or to support problem-solving by analyzing historical cases.
g) Handwriting Recognizers - It is supported by an expert system and neural computing
and is available in some-pen based computers. Scanners are good at reading typed or
printed material but they are not very good in handwriting recognition. Scanners can
interpret handwritten input and are subject to considerable error. Handwritten entries
input should follow very specific rules.
Application of Artificial Intelligence

Expert System - It is a software base system that attempts to capture human


knowledge or expertise in a limited domain area. It uses knowledge of human
experts represented in the form of heuristics. Heuristics is therefore a rule of
thumb or a rule of good guessing. It allows the expert system to function in a
manner consistent with human expert, advising the user on how to solve a
problem. Expert System captures expertise of knowledge in the form of rules,
which together with a database of relevant facts and information constitutes the
system knowledge-based. A real-world ES contains hundreds to thousands of
rules in an ES and tends to be far more interconnected, more numerous, and more
complex than traditional programs. ES can be inherently interactive, taking input
from user and coming up with the suggestions based on the stored knowledge.
Main Parts of Expert System
User Interface - Enable the manager to enter instructions and information
into ES and to receive information from IT. Instructions, It specify the
parameters that guide the ES through its reasoning process. Information,
it is in the form of value assigned to certain variables.
Knowledge Based - It contains the both the facts and the knowledge
representation techniques. Facts, it describes the problem area.
Knowledge Representation Techniques, it describes how the facts fit
together in a logical manner. Problem Domain, it describes the problem
area.
Inference Engine - This is the portion wherein performing reasoning by
using the content of the knowledge base on a particular sequence takes
place. During the consultation, this examines the rules of the knowledge
base one at a time. If the rule/condition is true, the specified action is
taken. For in ES terminology, the rule is fired when the action is taken.
Developmental Engine - This is used to create an Expert System. It
explains how the inference engine drew specific conclusion or why the
system is asking a specific question.

Neural Networks - It is a Computer Science that can simulate the functioning of


the brain. Whereas, if Expert System tries to represent human knowledge logically
through computer program, neutral net tries to represent the physical brain and
thinking process through electronic circuit or software.
Natural Language Processing (NLP) - It is concerned with understanding
natural languages like English, Spanish, and Filipino. Limited access in this area
is typified by systems that can recognize and interpret individual words or short
sentences. Examples of which are these:

Speech Recognition- area of AI whose goal is to develop machines which


can interpret and generate spoken words.
Speech Synthesis- gives the machine the ability to speak to users.
Robotics - It attempts to develop physical machines that can perform useful work
normally done by people. It has a sensory system which is vision system
combined with AI. The primary focus of robotics has been on factories and
hazardous environments, where machines can replace humans who would
otherwise be exposed to danger. ROBOT, it is an electromechanical device that
can performed and re-performed to automate manual tasks. Not all robotics is
considered part of AI, so to qualify as robotics the robot must:
Get information from its surroundings.
Do something physical, such as move or manipulate objects.

S-ar putea să vă placă și