An Introduction to
Computational Intelligent (CI)
Ismael A. Ali
Doctoral Seminar | Department of Computer Science
Kent State University | Spring 2014
iali1@kent.edu
Outline
 Motivation
 Main umbrella: Natural Computing
 Computational options: Levels of Abstraction
 Definition: CI
 Basic Properties of CI
 CI Main Paradigms
 Examples of Natural phenomenas
 Computational Intelligence: Modeling
Methodology
 Applications of CI
 Recommended References
 New world of computation:
 Mobility; computation in everyplace
 Dynamic; computation for everything
 Adaptation and improvement; computation in every-
environ
 Smartness: make cup of tea
 Uncertainty and Noise
 missing information
 Pattern recognition
 Needs for computation to survive:
 Think
 Adapt
 Sense
 Move
Why from nature?
- The processes are well done and successfully and
for years
- we see the “natural processes” as “information
processing systems”
- We are dealing with Interdisciplinary study
Main umbrella: Natural Computing
Do not confuse!
 Nature for computation: CI
 “Bio-computing”
 Computation for nature
 “Computational biology”
 Nature for engineering
 “Bio-inspired engineering”
Computational options:
Levels of Abstraction
Source: studyblue.com
Definition
 Computational intelligence (CI):
is a set of nature-inspired computational
methodologies and approaches
to address complex real-world problems to which
traditional approaches, i.e., first principles modeling
or explicit statistical modeling, are ineffective or
infeasible
wikipedi
a
Computational Intelligence
- Pseudonyms:
Natural Computation
Nature-Inspired Computing
Soft Computing
Adaptive Systems
AI vs. CI
CI is a specific subset of AI, but while AI focuses on
the outcome, CI focuses on the mechanism.
Deep Blue: highly optimized + extensive knowledge base
Basic Properties of CI
 Mobility
 Adaptability
 Complexity
 Dynamics
 Robustness
 Sustainability
 And more....
CI Main Paradigms:
 artificial neural networks
 evolutionary computation
 fuzzy logic
 swarm intelligence
 artificial immune systems
 ant algorithms
 bee algorithms
 and more to come!
Examples of Natural phenomenas
 Pattern recognition
 in different
 levels of abstraction
DNA
Neural Networks
Human Immune System
Swarm
Lets make an
Computational Intelligence
Algorithm !
Computational Intelligence
Modeling Methodology
Sample Modeling Methodology
Source: greensmith, DCA, PhD thesis
Applications of CI
• Air Traffic Control,
• Scheduling,
• Machine Learning,
• Pattern Recognition,
• Job Shop Scheduling,
• Earthquake Prediction
Systems,
• Market Forecasting,
• Data-Mining,
• User-Mining,
• Resource Allocation,
• Path Planning,
• System modeling,
• Network security
Recommended References
 Books
Recommended References
 Conferences
 IEEE Computational Intelligence Society:
http://cis.ieee.org/
 Courses
 CITS7212 Computational Intelligence: The
University of Western Australia
http://undergraduate.csse.uwa.edu.au/units/CITS7212/schedule.html
Go out to the nature and try made your own algorithm(s)!
Any Q?

Introduction to Computational Intelligent

  • 1.
    An Introduction to ComputationalIntelligent (CI) Ismael A. Ali Doctoral Seminar | Department of Computer Science Kent State University | Spring 2014 iali1@kent.edu
  • 3.
    Outline  Motivation  Mainumbrella: Natural Computing  Computational options: Levels of Abstraction  Definition: CI  Basic Properties of CI  CI Main Paradigms  Examples of Natural phenomenas  Computational Intelligence: Modeling Methodology  Applications of CI  Recommended References
  • 4.
     New worldof computation:  Mobility; computation in everyplace  Dynamic; computation for everything  Adaptation and improvement; computation in every- environ  Smartness: make cup of tea  Uncertainty and Noise  missing information  Pattern recognition  Needs for computation to survive:  Think  Adapt  Sense  Move
  • 5.
    Why from nature? -The processes are well done and successfully and for years - we see the “natural processes” as “information processing systems” - We are dealing with Interdisciplinary study
  • 6.
  • 7.
    Do not confuse! Nature for computation: CI  “Bio-computing”  Computation for nature  “Computational biology”  Nature for engineering  “Bio-inspired engineering”
  • 8.
  • 9.
  • 10.
    Definition  Computational intelligence(CI): is a set of nature-inspired computational methodologies and approaches to address complex real-world problems to which traditional approaches, i.e., first principles modeling or explicit statistical modeling, are ineffective or infeasible wikipedi a
  • 11.
    Computational Intelligence - Pseudonyms: NaturalComputation Nature-Inspired Computing Soft Computing Adaptive Systems
  • 12.
    AI vs. CI CIis a specific subset of AI, but while AI focuses on the outcome, CI focuses on the mechanism. Deep Blue: highly optimized + extensive knowledge base
  • 13.
    Basic Properties ofCI  Mobility  Adaptability  Complexity  Dynamics  Robustness  Sustainability  And more....
  • 14.
    CI Main Paradigms: artificial neural networks  evolutionary computation  fuzzy logic  swarm intelligence  artificial immune systems  ant algorithms  bee algorithms  and more to come!
  • 15.
    Examples of Naturalphenomenas  Pattern recognition  in different  levels of abstraction
  • 16.
  • 17.
  • 18.
  • 19.
  • 20.
    Lets make an ComputationalIntelligence Algorithm !
  • 21.
  • 22.
    Sample Modeling Methodology Source:greensmith, DCA, PhD thesis
  • 23.
    Applications of CI •Air Traffic Control, • Scheduling, • Machine Learning, • Pattern Recognition, • Job Shop Scheduling, • Earthquake Prediction Systems, • Market Forecasting, • Data-Mining, • User-Mining, • Resource Allocation, • Path Planning, • System modeling, • Network security
  • 24.
  • 26.
    Recommended References  Conferences IEEE Computational Intelligence Society: http://cis.ieee.org/  Courses  CITS7212 Computational Intelligence: The University of Western Australia http://undergraduate.csse.uwa.edu.au/units/CITS7212/schedule.html
  • 27.
    Go out tothe nature and try made your own algorithm(s)!
  • 29.