ARTIFICIAL INTELLIGENCE 1

ARTIFICIAL INTELLIGENCE

NAME

INSTITUTION AFFILIATED
 ARTIFICIAL INTELLIGENCE 2

ARTIFICIAL INTELLIGENCE

DEFINITION

According to Grewal (2014), Artificial Intelligence is the mechanical simulation system of

collecting knowledge and information and processing intelligence of universe: (collating and

interpreting) and disseminating it to the eligible in the form of actionable intelligence.

Artificial intelligence is also the science and engineering of making intelligent machines

as per McCarthy (1995)

REAL LIFE APPLICATIONS

Medicine; used in image guided surgery, image analysis and enhancement. Hence

enabling the medics to perform their tasks efficiently.

Transportation; used in autonomous vehicle control and pedestrian control. This has seen

the transport sector in developed countries grow and even easen traffic through pedetian and

traffic control

Bioinformatics; used in gene expression, data analysis, prediction of protein structure.

Research organizations have made use of artificial intelligence hence providing solutions to

problems.

Autonomous planning scheduling; used in autonomous rovers, telescope scheduling and

analysis of data.

Used in gaming especially in video games, in text classification and document sorting.
 ARTIFICIAL INTELLIGENCE 3

AGENTS IN A.I

An intelligent agent is a system that perceives its environment and takes actions which

maximize its chances of success as per Russel and Novig ( 2003)

Rational agent is an agent which acts in a way that is expected to maximize the

performance measure given the evidence provided by what it perceived and whatever built in

knowledge it has. The agents are as below;

Simple reflex agents are those that select an action based on the current state only,

ignoring history of perceptions. They only work if the environment is fully observable or the

correct action is based on what is perceived currently.

Goal based agents combine the provided goal information with the environment model to

choose the action which achieve the goal. It is an improvement of over model based and is used

in cases where knowing the current state of environment is not enough.

Utility based agent is that which will choose the action that maximizes the expected

utility. It is an improvement over goal based agents when achieving the desired goal is not

enough especially when considering cost.

Model based agents keep track of partially observable environment. They have an

internal state depending on perception history, the environment is modelled based on how it

evolves independently from the agent and how the agent actions affect the world.
 ARTIFICIAL INTELLIGENCE 4

ALGORITHMS IN A.I

Breadth first search algorithm searches breadthwise starting from the root node

expanding to all successor nodes at the current level before moving to the next level. It uses the

FIFO queue data structure.it provides a solution with minimal steps if the solution exists but it

requires a lot of memory as each level is saved and it takes a lot of time if the solution is far from

the root node.

Depth first search is a recursive algorithm for traversing a tree or graph. It starts from the

root node and follows each path to its greatest depth before moving to the next path. It uses stack

data structure for implementation. It uses less memory and less time to find a solution but

recursion may lead to no guarantee of reaching the s olution and may lead to an infinite loop.

Completeness: is the strategy guaranteed to find a solution if there is one.

Time complexity: how long does it take to find a solution.

Space complexity: how much memory does it take.

Optimality: how does the strategy find the highest quality solution when there are several

different solutions.
 ARTIFICIAL INTELLIGENCE 5

EXPERT SYSTEM AND ITS ARCHITECHTURE

An expert system is an intelligent computer program that uses knowledge and reasoning

procedures to solve difficult problems that need certain expertise to be solved. It was build to

replace and help an expert

 An expert system compared with traditional computer:

ARCHITECTURE

Knowledge base; it stores facts and rules, it does not store information like texts, numbers

and logical values. It stores concepts and dedicated procedures that need to be done to solve a

problem. methods of storing knowledge in database are predicate calculus, mainframe and

semantic network.

Knowledge acquisition; Is a process of gathering and transferring problem solving

expertise from all source of knowledge in a computer program, this information is used to

develop and expand the knowledge base.

Inference engine; Is a computer program that drives to the conclusion and provide the

reasoning methodology for information stored in knowledge base. It also provides guideline on

using the knowledge in expert system.it uses two strategies that is forward chaining and

backward chaining.
 ARTIFICIAL INTELLIGENCE 6

Explanation facility; This component acts to help the user understand how an expert

system reaches a certain decision of the problem to be solved.

a) User interface; It enables the user to query the system input

information and receive advice.

b) Working memory; it contains facts of problems that are happening

during consultation process with the expert system.

STRATEGIES IN INFERENCE ENGINE

Backward chaining starts from the decision and moves backward to obtain supporting

facts for the decisions made. If there are no matching facts that supports the decision, the

decision will be rejected and another will be chosen. the process continues until a suitable

decision and the facts that support it are obtained.

Forward chaining starts reasoning from facts provided and moves on until it achieves

decision, it is guided by the facts in memory space and premises which it can obtain tem from.

Engine tries to

match the required premise(if) for all rules in the knowledge database with the facts given

which are in its memory, if there are several rules that match the solving procedures will be used,

it will do this repeatedly.

PREDICATE AND PROPOSITION LOGIC

Predicate logic is a form of logic that contains variables (predicate variables) and may be

true or false depending on the values of these variables. It is an extension of propositional logic
 ARTIFICIAL INTELLIGENCE 7

while propositional logic is a form of logic where all statements are made by declarative

statements which are either true or false.

CONFLICT RESOLUTION STRATEGIES

First in first serve; The rule applied will be the first rule that is matched.

Last in first serve; The rule applied will be the last rule that is matched.

Prioritization; The rule to apply will be based on priorities set on rules, with priority

information usually provided by an expert or knowledge engineer.

Specificity; The rule applied is usually the most specific rule, or the rule that matches the

most facts. This strategy cannot be used if the rules are equal in specificity.

Recency; The rule applied is the rule that matches the most recently derived facts

Fired rules; Involves not applying rules that have already beedn used .

Line of reasoning; Involves firing all ruels with a separate line of reasoning.

ALGORITHMS IN MACHINE LEARNING

Supervised algorithm is a type of machine learning in which machines are trained using

labelled training data and on basis of that data, machines predict the output. The aim is to find a

mapping function to map the input variable(x) with the output variable(y)

Unsupervised algorithm is a type of machine learning in which models are trained using

unlabeled dataset and are allowed to act on that data without any supervision. The goal is to find
 ARTIFICIAL INTELLIGENCE 8

the underlying structure of dataset, group that data according to similarities and represent that

dataset in a compressed format.

Supervised algorithm models are trained using labelled data ,the models predict the

output and the take direct feedback to check if its predicting correct output or not.

For example;if we have an image of different types of fruits. The task of our supervised

learning model is to identify the fruits and classify them accordingly. So to identify the image in

supervised learning, we will give the input data as well as output for that, which means we will

train the model by the shape, size, color, and taste of each fruit. Once the training is completed,

we will test the model by giving the new set of fruit. The model will identify the fruit and predict

the output using a suitable algorithm.

Unsupervised algorithm models are trained using unlabeled data, they find the hidden patterns in

data and does not take feedback.

For example; If the unsupervised learning algorithm is given an input dataset containing

images of different types of cats and dogs. The algorithm is never trained upon the given dataset,

which means it does not have any idea about the features of the dataset. The task of the

unsupervised learning algorithm is to identify the image features on their own. Unsupervised

learning algorithm will perform this task by clustering the image dataset into the groups

according to similarities between images. Predicate logic is a form of logic that contains

variables (predicate variables) and may be true or false depending on the values of these

variables. It is an extension of propositional logic while propositional logic is a form of logic

where all statements are made by declarative statements which are either true or false.
 ARTIFICIAL INTELLIGENCE 9

DEFINITION OF TERMS

Dataset is a collection of data in which data is arranged in some order. It contains any

data from a series of an array to a database.

Training set is a subset to train a model, dataset of examples used during the learning

process and is used to fit parameters.

Test set is a subset to test a trained model, is a dataset used to provide an unbiased

evaluation of final model fit on the training dataset.

Cross validation is a resampling procedure used to evaluate machine learning models on

a limited data sample.

Confusion matrix is a table that is often used to describe the performance of a

classification model or a classifier on a set of test data for which the true values are known.
 ARTIFICIAL INTELLIGENCE 10

REFERENCES

Andresen, S. L. (2002). John McCarthy: father of AI. IEEE Intelligent Systems, 17(5), 84-85.

Grewal, D. S. (2014). A critical conceptual analysis of definitions of artificial intelligence as

applicable to computer engineering. IOSR Journal of Computer

Engineering, 16(2), 9-13.

Livingstone, D. J. (Ed.). (2008). Artificial neural networks: methods and applications (pp. 185-

202). Totowa, NJ, USA: Humana Press.