Specialized Business Information Systems

1
Specialized Business Information Systems

• Chapter 7

2
The Nature of Intelligence

• Learn from experience apply the knowledge
• Deep Blue improves its performance by playing
  with humans
• Handle complex situations
• Traffic problem in Istanbul
• Solve problems when important information is
  missing
• Based on available information
• Determine what is important
• Choose which facts to use to compute the solution

3
The Nature of Intelligence

• React quickly and correctly to new situations
• Requires understanding the new situation
• Understand visual images
• Requires perception
• Process and manipulate symbols
• Computers are better at dealing with numbers
• Be creative and imaginative
• Use heuristics
• Rules of thumb from experience

4
A comparison of Natural and Artificial
Intelligence
5
A Conceptual Model of Artificial Intelligence
6
What is an Expert System?

• Hardware and software that contain knowledge and
  manipulate knowledge by inferences
• Mycin (Shortliffe 1976) Expert system for
  medicine
• Program for advising physicians on treating
  bacterial infections
• Question and answer dialogues with user
• Accounts for uncertainties
• Explains its reasoning

7
Characteristics of an Expert System

• Can explain their reasoning or suggested
  decisions
• Why recommend a certain medicine?
• Can display intelligent behavior
• Can draw conclusions from complex relationships
• A patient is diagnosed with two diseases,
• The cures for the diseases may have conflicts
• Can provide portable knowledge
• Capture knowledge in ones brain
• Can deal with uncertainty
• A patient is diagnose without running all the
  tests

8
Characteristics of an Expert System

• Not widely used or tested
• Limited to relatively narrow problems
• Cannot readily deal with mixed knowledge
• Expert systems should talk to each other
• Cannot refine its own knowledge
• Should be able to keep a consistent knowledgebase
• Should have a way to gain new knowledge
• May have high development costs
• Raise legal and ethical concerns

9
When to Use Expert Systems

• High payoff
• Preserve scarce expertise
• Provide more consistency than humans
• Faster solutions than humans
• Training expertise

10
Components of an Expert System
11
The Relationships Among Data, Information, and
Knowledge
12
Rules for a Credit Application
13
The Knowledge Acquisition Facility
14
Knowledge Base

• Assembling human experts
• Combine knowledge from several experts
• Disagree on many items
• The use of fuzzy logic
• For relations that are not precise
• Is a 50-year old man old?
• Help computers deal with imprecise knowledge
• Ex Washing machines Auto-focus cameras

15
Knowledge Base

• The use of rules
• Rule Conditional statement (if then)
• If the condition matches, the action fires
• More rules generally mean more precision
• The use of cases
• Template of problems or situations
• To find the solution of a new case, find similar
  old cases and apply result

16
Inference Engine (1)

• Use information and relations to derive new facts
  to solve problems or predict possible outcomes
• Main reasoning component
• Find the right facts, apply the right relations,
  etc.
• Ex Facts male(Ali), female(Oya)
• Relations father(X, Y) gt male(X)
• The engine can conclude that Oya cannot be a
  father.

17
Inference Engine (2)

• Backward chaining
• You start with conclusions
• You want to find out if you can get to the
  conclusion from your facts
• Forward chaining
• You start with facts and try to reach conclusions
• More expensive since it can generate many
  conclusions

18
Other Components

• Explanation Facility
• Enables the expert system to explain its
  reasoning
• Helps the user to judge the expert system
• Knowledge Acquisition Facility
• Get and update knowledge
• Provide a way to capture and store knowledge
• Can be semi-automated
• User Interface
• Help users interact with the system
• Improve usability

19
Expert Systems Development
20
Participants in Developing and Using Expert
Systems
21
Domain Expert

• Recognize the real problem
• Develop a general framework for problem solving
• Formulate theories about the situation
• Develop and use general rules to solve a problem
• Know when to break the rules or general
  principles
• Solve problems quickly and efficiently

22
Other participants

• Knowledge Engineer
• Works in design and implementation of the expert
  system
• Has considerable information about expert systems
• Knowledge User
• End user who will benefit from the system
• No need to know anything about expert systems
• Can help in testing

23
Expert Systems Development Alternatives
24
Applications of Expert System and Artificial
Intelligence

• Credit granting and loan analysis
• Stock Picking
• Catching cheats and terrorists
• NORA (Non-obvious Relationship Awareness)
• Budgeting

25
Applications of Expert System and Artificial
Intelligence

• Games Proverb solves crossword puzzles
• Writing Evaluate and rate writings
• Information management and retrieval
• Virus detection
• Learns the actions of a virus
• Hospitals and medical facilities

26
Virtual Reality

• Enables one or more users to move and react in a
  computer-simulated environment
• Immersive virtual reality - user becomes fully
  immersed in an artificial, three-dimensional
  world that is completely generated by a computer
• Virtual reality system - enables one or more
  users to move and react in a computer-simulated
  environment

27
Interface Devices

• Head mounted display (HMD)
• Binocular Omni-Orientation Monitor (BOOM)
• CAVE

28
The BOOM, a Head-Coupled Display Device
29
Viewing the Detroit Midfield Terminal in an
Immersive CAVE System
30
Useful Applications

• Medicine used to link stroke patients to
  physical therapists
• Education and training used by military for
  aircraft maintenance
• Entertainment
• Star Wars Episode II Attack of the Clones
• Real Estate Marketing and Tourism
• Used to increase real estate sales
• Virtual reality tour of the White House

31
Segway
32
Robotics

• Mechanical or computer devices that can move
  autonomously
• Manufacturers use robots to assemble or paint
  products
• Asimo in Istanbul Shake hands, dance
• Unmanned Combat Air Vehicles (UCAVs) Identify
  and destroy targets without human intervention

33
Vision Systems

• Capture, store, manipulate visual images
• Fingerprint analysis Store a database of
  fingerprints and information about the owners.
• Match a fingerprint with an existing entry in the
  database
• Mostly recognize black and white

34
Natural Language Processing

• Understand and react to statements in natural
  language
• Three levels of understanding
• Commands
• Discrete
• Continuous
• Talk to a computer computer converts languages
  to commands understandable by computers

35
Learning Systems

• Change its behavior over time
• Computer takes an action
• User gives feedback
• Based on the feedback, computer modifies its
  action
• First train the system then try on test data
• Amazon.com learns user models as users browse and
  buy goods

36
Summary

• Artificial intelligence - used to describe
  computers with ability to mimic or duplicate
  functions of the human brain
• Intelligent behavior - includes the ability to
  learn from experience
• Expert systems - can explain their reasoning (or
  suggested decisions) and display intelligent
  behavior
• Virtual reality system - enables one or more
  users to move and react in a computer-simulated
  environment
• Special-purpose systems - assist organizations
  and individuals in new and exciting ways

37
Principles and Learning Objectives

• Artificial intelligence systems form a broad and
  diverse set of systems that can replicate human
  decision making for certain types of well-defined
  problems.
• Define the term artificial intelligence and state
  the objective of developing artificial
  intelligence systems.
• List the characteristics of intelligent behavior
  and compare the performance of natural and
  artificial intelligence systems for each of these
  characteristics.
• Identify the major components of the artificial
  intelligence field and provide one example of
  each type of system.

38
Principles and Learning Objectives

• Expert systems can enable a novice to perform at
  the level of an expert but must be developed and
  maintained very carefully.
• List the characteristics and basic components of
  expert systems.
• Outline and briefly explain the steps for
  developing an expert system.
• Identify the benefits associated with the use of
  expert systems.

39
Principles and Learning Objectives

• Virtual reality systems have the potential to
  reshape the interface between people and
  information technology by offering new ways to
  communicate information creatively.
• Define the term virtual reality and provide three
  examples of virtual reality applications.
• Special-purpose systems can help organizations
  and individuals achieve their goals.
• Discuss examples of special-purpose systems for
  organizational and individual use.