Artificial intelligence

Artificial Intelligence

Mr Arthur

Aims of Lesson 1

1. What is Intelligence??2. Turing Test3. Problems with the Turing Test

What is Intelligence????

• “Intelligence is the name we give to the data processing activity of entities which respond to information with behaviour which appears to be intended to be optional with respect to pre-set goals”

• “Intelligence involves Knowing and Choosing”

Intelligence??

Many researchers feel that intelligence has the following features The Ability to:

Learn and adapt from experience. Retain knowledge and use to make decisions. Problem solve. Handle and manipulate language

“Artificial intelligence is concerned with building machines that can act and react appropriately, adapting their response to the demands of situation.”

Testing Intelligence

Turing Test Alan Turing, a British mathematician developed a test

in 1950 to determine if a program was intelligent: It has the following features:

A human tester was connected to 2 terminals and asked a series of questions.

One of the terminals had another human making the responses and the other used a computer program to make the responses

If the human tester could not distinguish between the human responder and the program response then the software was said to be intelligent.

Problems with Turing Test

It needed a fairly limited problem domain i.e. the area of the knowledge had to be small.

Did the program pretend to forget things or make errors in order to fool the human tester?

Lesson Starter

1. Give 3 features of Intelligence

2. Give 3 examples of AI in everyday life

3. What is the name of the test that is used to determine if a system is Intelligent?

4. Describe this test

Aims of Lesson 2

1. Early AI Developments – Game Playing2. Chatterbots

1. ELIZA2. SHRDLU3. Parry

Early Developments in AI (1940-65)

In the beginning the focus of AI research was on modelling the human brain. (This was impossible)

Research shifted to using games like noughts and crosses, drafts etc to create “AI” systems The games had a number of rules that

were easy to define.

Language Processing (1965-1975)

In 1965 Researchers agreed that game playing programs could not pass the Turing test

The focus shifted to language processing ELIZA (1966)

1st language processing program Responded to users inputs by asking questions

based on previous responses

Language Processing (1965-1975)

PARRY (1972) Parry modelled a conversation with a

paranoid person This seems odd but the program was

created by a psychiatrist SHRDLU (1970)

The program could interpret verbal commands to move coloured blocks

"Move the red block behind the green one" It could understand and carry out the

instruction

Lesson Starter

1. Give 1 problem with the Turing test

2. Why did early AI development focus on Game Playing?

3. Describe the response from the chatterbot ELIZA

Aims of Lesson 4

Last Lesson Definition of intelligence The Turing Test Early Developments in AI =

Game Playing Language Processing

1. ELIZA2. SHRDLU3. PARRY4. Chatterbots

Today’s Lesson Developments in

Hardware

Developments in Hardware

Faster processors More instructions can be processed per second

More RAM memory Larger knowledge bases can be open and stored in RAM

Increased backing storage capacity Larger knowledge bases can be stored and reopened

Multiple processors (Parallel) Where different processors can be processing different

instructions

Aims of Lesson 5




Today’s Lesson Expert Systems

Expert Systems

An expert system is a computer program which uses a knowledge base of facts and rules populated by a human expert

The expert system will give reliable advice on a limited area of expertise, and can interact and explain it reasoning to the user (justification facilities)

Examples NHS 24 MYCIN to diagnose blood disorders Legal advice Chemical analysis, DENDRAL was designed to identify unknown

substances Car mechanic expert system BABY used to monitor premature babies

Aims of Lesson 5




Today’s Lesson Expert Systems

Advantages/Disadvantages

Advantages of Expert Systems

Expertise Available to access 24/7 Reduced wage bill Expert System will never get tired You can have multiple copies of the expert system Can combine the knowledge of many experts No human emotion involved No possibility of expert system retiring No barrier due to poor communication skills

Social/Legal/Ethical Issues of Expert Systems

No common sense is used Legal = If the expert system makes an error

who is responsible? Developer? Company using it?

Moral = Expert system wouldn't make a decision on what is morally correct, it would only use facts and rules

Loss of jobs Loss of human expertise over time

Aims of Lesson 6




Expert Systems Advantage/Disadvantage

Today’s Lesson Artificial Neural Systems Adv/Dis

Revision Questions1. Give 4 features of Intelligence

2. I have a black and white image that is 3 inches by 7 inches with a Resolution of 400 dpi. Calculate the storage requirements in Kb

3. Convert 152 to binary

4. Describe a Client Server network

5. List the 7 stages in the SD process

6. What is a Macro?

7. Give 3 symptoms of your computer being infected by a virus

8. Give 3 features that you could use to Evaluate software

9. Give 3 disadvantages of an Expert System

10. How many bits are used to represent a character is ASCII?

Artificial Neural Systems

ANS is an approach to AI where the developer attempts to model the human brain

Simple processors are interconnected in a way that simulates the connection of nerve cells in the brain

The output from the ANS is compared with the expected output and the processors can be “retrained”

1. Give 3 features of intelligence

2. What is the name of the Test that is used to test intelligence?

3. Describe this test

4. Why did early AI development focus on Game Playing?

5. Give 3 examples of chatterbots

6. Describe the response the user would get from ELIZA

7. What is an Expert System?

8. Give 2 examples of Expert Systems

9. Give 2 advantages of Expert Systems

10. Give 2 possible disadvantages of an Expert System

11. What is meant by the term Artifical Neural System

Applications of ANS

Post Office has been using ANS to automate the reading of postcodes

An ANS system to predict the stock market Assessing debt risk of an individual

Adv/Disadv of ANS

Advantages They can learn without

needing to be reprogrammed

They have a good success rate at predicting the correct response

Disadvantages Time consuming and

requires a lot of technical expertise to set up

ANS cannot explain reasoning behind decision

Lesson Starter

1. What is an Artificial Neural System?

2. Give 2 advances in hardware that has led to the development of AI

3. What is meant by the term parallel processing?

Aims of Lesson 6





Artificial Neural Systems Adv/Dis

Today’s Lesson Vision Systems

Vision Systems AI can be used to recognise and make sense of

images. Uses

Security systems, recognising faces at airports Inspection of manufactured goods judging quality of

production Vision systems on automated cars Interpretation of Satellite photos for military use

Stages1. Input Image using Digital Camera2. Detect Edges of Object3. Compare to Knowledge Base – Pattern Matching4. Understanding Object

Vision Systems

Difficulties with Vision Systems Shadows on Objects Identifying the Edge of the Image Glare Objects hiding other parts of the Image Viewing from different angles

Lesson Starter

1. Give 3 examples of Vision Systems in everyday life

2. Give the 4 stages in a Vision System

3. Give 2 possible problems with Vision Systems

Aims of Lesson 8






Vision Systems

Today’s Lesson Natural Language

Processing

Natural Language Processing

NLP or Speech Recognition is where an AI system can be controlled and repsond to verbal commands

Examples Speech-driven word processors Computer operation for disabled users Military weapon control Mobile phones Customer query lines

Problems with NLP

The need to train the NLP system to your voice Background noise Ambiguity of words and phrases

I saw a man eating fish I saw a man hitting a boy with a stick

Changing English language i.e. bouncebackability Accents and dialects i.e. I ken what you are talking

about!! Similar sounding words and phrases e.g. furry boots are

you fae?? Inconsistencies with grammar

Lesson Starter

1. What is meant by the term Natural Language Processing?

2. Give 3 examples in everyday life

3. Give 3 problems with accuracy of NLP systems

Aims of Lesson 9






Vision Systems Natural Language Processing

Today’s Lesson Hand-writing recognition

Handwriting Recognition

This is where handwritten words are converted into editable text

Applications Mainly used for Input into Palmtop computers

and Tablet PCs Hand writing varies considerably so there

is a need to train the software to your style of handwriting

Aims of Lesson 10






Vision Systems Natural Language Processing

Today’s Lesson Intelligent Robots

Intelligent Robots Robots can be considered intelligent when they go

beyond simple sensors and feedback (dumb robots), and display some further aspect of human-like behaviour

Vision Systems The ability to learn and improve performance Robot that can walk rather than on wheels NLP response

Examples The delivery of goods in warehouses The inspection of pipes Bomb Disposal Exploration of Ocean floor or space

Advantages of Intelligent Robots

Can be used in dangerous environments More accurate than humans No wages or holidays Work 24/7 Don’t need to be constantly programmed

Aims of Lesson 11Last Lesson Definition of intelligence The Turing Test Early Developments in AI =





Vision Systems Natural Language Processing Intelligent Robots

Today’s Lesson Knowledge Representation

Semantic Nets Facts, Rules and Queries

Knowledge Representation

AI programs cannot solve a problem without storing information about the problem in a Knowledge Base

Semantic nets are used to represent known facts before coding the Knowledge Base

is a (tiger, cat)is a (lion, cat)eats(tiger, meat)eats(lion, meat)has(tiger, stripes)is a (deer, meat)is a (zebra, meat)eats(tiger, meat)eats(lion, meat)

Semantic Nets

Mint and basil are herbs.

Pennyroyal and spearmint are types of mint.

Genovese basil and purple basil are two types

of basil.

Mints can be used in making tea, basil is used

In the making of pesto. (6)

Semantic Net 2

A fish has fins, scales, is cold blooded and has

gills. There are lots of kinds of fish, including

cod, haddock, shark, goldfish, and trout. A fish

is a vertebrate - that means it has a backbone. I

have a goldfish called Sammy. Cod, haddock

and sharks are salt-water fish (they live in salt

water). Goldfish and trout are fresh-water fish

(they live in fresh water).

The programming language Prolog is used to define facts, and rules in a knowledge base that can be queried to solve the problem

Facts isa(nicole, female). isa(ricky, male). likes(elena, spaceraiders).

Rules cheesy(X):-

isa(X,male),

likes(X, wotsits).

Queries ?likes(X,monstermunch). ?isa(X, male). ?isa(jake,female). ?cheesy(X).

Aims of Lesson 12Last Lesson Definition of intelligence The Turing Test Early Developments in AI = Game Playing Language Processing




Vision Systems Natural Language Processing Intelligent Robots Knowledge Representation

Semantic Nets Facts, Rules and Queries

Today’s Lesson Search techniques

Search Techniques

A search tree allows you to show the relationships between information.

Search Techniques Depth First Search

This search starts at the top most node and travels as far down the left hand “branch” as far as it can go.

A B D B E B A C F C G

Breadth First Search A breadth first search visits each child node in turn.

It works its way across each level. A B C D E F G

Arrangements

The development of artificial intelligence Description of human intelligence (including the ability to communicate, retain

knowledge, solve problems) Description of the Turing test and explanation of its rationale Explanation of the need for a different approach to programming which could

represent knowledge Simple description of the development of game playing programs from simple

early examples to contemporary complex examples exhibiting intelligence Simple description of the development of language processing from Eliza to

chatterbots and contemporary applications Simple description of the development of expert systems Identification of hardware developments (including faster processors, more

memory, and increasing backing store capacity) which have assisted the development of AI

ArrangementsExpert systems: Description of purpose of expert systems Description of advantages of expert systems over human experts, including expertise always available, reduced wage bill, combines

expertise of several experts, less chance of errors Description of contemporary applications of expert systems Description of social, legal and ethical issues related to the use of expert systems (including loss of jobs, training issues, public

reactions, loss of human expertise)

Artificial neural systems: Simple description of a neural network as an electronic model of the brain consisting of many interconnected simple processors Description of uses and examples of artificial neural systems (including learning to read postcodes; stock market prediction; debt

risk assessment; other examples of pattern recognition ) Description of advantages and disadvantages of artificial neural systems

Vision systems: Explanation of the need to interpret/make sense of visual input. Description of applications (including industrial, military use, satellite photo interpretation)

Speech recognition: Description of applications (including word processor, punctuation commands, disabled users, cars, military, mobile phones) Description of characteristics (training for each voice pattern, control instructions, influence of background noise, factors affecting

accuracy) Handwriting recognition: Description of common applications (including palmtops and tablet PCs) Explanation of possible need to train the system

ArrangementsSearch techniques Exemplification of problem solving by search Construction of a simple search tree Description of breadth-first and depth-first search and exemplification on a search tree

Knowledge representation Construction of semantic net to represent simple relationships and facts Description and exemplification of the following features in Prolog (or similar declarative

language): simple facts (single/double argument), simple rules (up to two sub-goals), simple queries (true/false, single variable), operators: and, >, < , =

Explanation of the concepts of goals and sub-goals Perform simple manual trace: one rule/level

Intelligent robots: Description of types of sensors used Description of contemporary applications (including automated delivery, pipe inspection, bomb

disposal, exploration of unknown environments) Description of advantages of intelligent robots

Artificial intelligence

Education

reduced wage

artificial

larger knowledge

natural language

game playing

game playing

language processing

expert systems