1. TECHNOLOGY GUIDE 4: Intelligent Systems

2. TECHNOLOGY GUIDE 4: INTELLIGENT SYSTEMS
TG 4.1 Introduction to Intelligent Systems TG 4.2 Expert Systems TG 4.3 Neural Networks TG 4.4 Fuzzy Logic TG 4.5 Genetic Algorithms TG 4.6 Intelligent Agents
Copyright John Wiley & Sons Canada

3. Copyright John Wiley & Sons Canada
INTELLIGENT SYSTEMS
TG 4.1 Introduction to Intelligent Systems TG 4.2 Expert Systems TG 4.3 Neural Networks TG 4.4 Fuzzy Logic TG 4.5 Genetic Algorithms TG 4.6 Intelligent Agents
Copyright John Wiley & Sons Canada

4. Copyright John Wiley & Sons Canada
LEARNING OBJECTIVES
Differentiate between artificial intelligence and human intelligence.
Define “expert systems,” and provide examples of their use.
Define “neural networks,” and provide examples of their use.
Copyright John Wiley & Sons Canada

5. LEARNING OBJECTIVES (CONTINUED)
Define “fuzzy logic,” and provide examples of its use.
Define “genetic algorithms,” and provide examples of their use.
Define “intelligent agents,” and provide examples of their use.
Copyright John Wiley & Sons Canada

6. TG 4.1 INTRODUCTION TO INTELLIGENT SYSTEMS
Intelligent systems: information systems that can make decisions by themselves.
Examples: Web apps and medical uses
Major categories of intelligent systems:
expert systems
neural networks
fuzzy logic
genetic algorithms
intelligent agents
Intelligent systems describes the various commercial applications of artificial intelligence.
Artificial intelligence (AI) is a subfield of computer science that studies the thought processes of humans and recreates the effects of those processes via machines, such as computers and robots.
Click on the links in this slide to a website on an app to make your own music and to see a “virtual brain”
Copyright John Wiley & Sons Canada

7. NATURAL VS. ARTIFICIAL INTELLIGENCE
The following capabilities are considered to be signs of human intelligence: learning or understanding from experience, making sense of ambiguous or contradictory messages, and responding quickly and successfully to new situations.
The ultimate goal of AI is to build machines that mimic human intelligence.
Copyright John Wiley & Sons Canada

8. Copyright John Wiley & Sons Canada
TG 4.2 EXPERT SYSTEMS
Click here to access the Website of IBM Watson Supercomputer
Expert systems (ESs) are computer systems that attempt to mimic human experts by applying expertise in a specific domain. Expert systems can either support decision makers or completely replace them. An ES typically is decision-making software that can perform at a level comparable to a human expert in certain specialized problem areas.
Since 2007, IBM scientists have been trying to automate one of the most human of abilities: answering questions asked in everyday language, or natural language. They even gave their technology a human name: Watson. Watson uses more than 100 algorithms to analyze a question in different ways, providing hundreds of possible answers.
Copyright John Wiley & Sons Canada

9. EXPERTISE TRANSFER FROM HUMAN TO COMPUTER
Knowledge acquisition
Knowledge representation
Knowledge inferencing
Knowledge transfer
Knowledge acquisition: acquired from domain experts or from documented sources.
Knowledge representation: organized as rules or frames (objective-oriented) and stored electronically in a knowledge base.
Knowledge inferencing: the computer is programmed so that it can make inferences based on the stored knowledge.
Knowledge transfer: inferenced expertise is transferred to the user in the form of a recommendation.
Copyright John Wiley & Sons Canada

10. THE COMPONENTS OF EXPERT SYSTEMS
Knowledge base
Inference engine
User interface
Blackboard (workplace)
Explanation subsystem (justifier)
Knowledge base contains knowledge necessary for understanding, formulating and solving problems comprised of facts & rules.
Inference engine is a computer program that provides a methodology for reasoning and formulating conclusions.
User interface enables users to communicate with the computer
Blackboard is an area of working memory set aside for the description of a current problem.
Explanation subsystem explains its recommendations.
Copyright John Wiley & Sons Canada

11. FIGURE TG 4.1 STRUCTURE AND PROCESS OF AN EXPERT SYSTEM
Copyright John Wiley & Sons Canada

12. TABLE TG 4.2 TEN GENERIC CATEGORIES OF EXPERT SYSTEMS
Copyright John Wiley & Sons Canada

13. TABLE TG 4.3 BENEFITS OF EXPERT SYSTEMS
Copyright John Wiley & Sons Canada

14. DIFFICULTIES OF USING ES
Difficulty transferring domain expertise from human experts to the expert system
Challenge to automate certain processes
Potential liability
Copyright John Wiley & Sons Canada

15. Copyright John Wiley & Sons Canada
TG 4.3 NEURAL NETWORKS
Neural network is a system of programs and data structures that simulates the underlying concepts of the biological brain.
This slide illustrates how a neural network might process a typical mortgage application. Note that the network has three levels of interconnected nodes (similar to the human brain): an input layer; a middle, or hidden, layer; and an output layer. When the neural network is trained, the strengths, or weights, of its connections change.
Copyright John Wiley & Sons Canada

16. Copyright John Wiley & Sons Canada
TG 4.4 FUZZY LOGIC
Fuzzy logic is a branch of mathematics that deals with uncertainties by simulating the processes of human reasoning.
Examples:
Financial analysis (loan application)
Accounting (goodwill)
Internet searches (search queries)
The rationale behind fuzzy logic is that decision making is not always a matter of black or white, or true or false: It often involves grey areas where the term maybe is more appropriate.
Copyright John Wiley & Sons Canada

17. Copyright John Wiley & Sons Canada
TG 4.5 GENETIC ALGORITHMS
Genetic algorithms have three functional characteristics:
Selection (survival of the fittest): Giving preference to better and better outcomes.
Crossover: Combining portions of good outcomes in the hope of creating an even better outcome.
Mutation: Randomly trying combinations and evaluating the success (or failure) of an outcome.
Genetic algorithm mimics the evolutionary, “survival-of-the-fittest” process to generate increasingly better solutions to a problem. That is, a genetic algorithm is an optimizing method that finds the combination of inputs that produces the best outputs.
Copyright John Wiley & Sons Canada

18. Copyright John Wiley & Sons Canada
TG 4.6 INTELLIGENT AGENTS
Three types of Intelligent Agents (also called bots):
Information Agents
Monitoring-and-Surveillance Agents
User Agents
Information agents search for information and display it to users.
Monitoring and surveillance agents constantly observe and report on some item of interest.
User agents take action on a user’s behalf.
Copyright John Wiley & Sons Canada

19. INTELLIGENT AGENTS CONTINUED
Information agents search for information and display it to users.
Monitoring-and-surveillance agents, also called predictive agents, constantly observe and report on some item of interest.
User agents, also called personal agents, take action on your behalf.
A buyer agent, also called a shopping bot, helps customers find the products and services they need on a website.
Click on the links in this slide to access examples of intelligent agents.
Copyright John Wiley & Sons Canada

20. TECHNOLOGY GUIDE CLOSING
There are a number of characteristics that differentiate artificial and human intelligence.
Expert systems are computer systems that attempt to mimic human experts by applying expertise in a specific domain.
A neural network is a system of programs and data structures that simulate the underlying concepts of the human brain.
Copyright John Wiley & Sons Canada

21. TECHNOLOGY GUIDE CLOSING (CONTINUED)
Fuzzy logic is a branch of mathematics that deals with uncertainties by simulating the processes of human reasoning.
A genetic algorithm is an intelligent system that mimics the evolutionary, “survival-of-the-fittest” process to generate increasingly better solutions to a problem.
An intelligent agent is a software program that assists you, or acts on your behalf, in performing repetitive, computer-related tasks.
Copyright John Wiley & Sons Canada

22. Copyright John Wiley & Sons Canada