1 of 27

Rizoan Toufiq

CSE 4203�Introduction: Course Outline

Assistant Professor

Department of Computer Science & Engineering

Rajshahi University of Engineering & Technology

rizoantoufiq@cse.ruet.ac.bd

Course Title: Neural Networks and Fuzzy Systems

2 of 27

Basic

Course No: CSE 4203
Course title: Neural Networks and Fuzzy Systems
Prerequisite courses: N/A
Contact hours/week: 3
Credits: 3.00

2

Rizoan Toufiq, Assistant Prof, CSE, RUET

3 of 27

Course Description

Introductory Concept:

Introduction Human Brain Mechanism,
Neural Machine Intelligence.

Fundamental Concept of Neural Network:

Basic Models of Artificial Neuron,
Activation Function,
Network Architecture,
Neural Network Viewed as Directed Graph,
Basic Learning Rules,
Overview of Perceptron's,
Single Layer of Perceptron's,

3

Rizoan Toufiq, Assistant Prof, CSE, RUET

4 of 27

Course Description

Fundamental Concept of Neural Network:

Mathematical Model of Single Layer Perceptron's,
Perceptron's Learning Algorithm,
Delta Learning Rule,
Multi-Layer Perceptron's,
Back Propagation Learning Algorithm,
Mathematical Model of MLP Network.

Function Approximation:

Basis Function Network,
Radial Basis Function Networks (RBF),
MLP vs. RBF Networks,
Support Vector Machine (SVM).

4

Rizoan Toufiq, Assistant Prof, CSE, RUET

5 of 27

Course Description

Competitive Network and Associative Memory Network:

Adaptive Resonance Theory (ART),
ART-1 Architecture and Algorithm,
Kohonen Self-Organizing Maps (SOMs),
Linear Feed-Forward Associative Memory Network,
Recurrent Associative Memory Network,
Bidirectional Associative Memory Network (BAM),
Hopfield Networks.

5

Rizoan Toufiq, Assistant Prof, CSE, RUET

6 of 27

Course Description

Fuzzy System:

Introduction to Fuzzy System,
Fuzzy Relations,
Fuzzy Numbers,
Linguistic Description and their Analytical Form,
Fuzzy Control.

Defuzzification:

Defuzzification Methods,
Centroid Method,
Center of Sum Method,
Mean of Maxima Defuzzification,
Applications,

6

Rizoan Toufiq, Assistant Prof, CSE, RUET

7 of 27

Course Description

Defuzzification:

Equilibrium of Learning System,
Concept of Neuro-Fuzzy and Neuro-GA Network.

Genetic Algorithm:

Basic Concepts,
Offspring,
Encoding,
Reproduction,
Crossover,
Mutation Operator,
Application of GA.

7

Rizoan Toufiq, Assistant Prof, CSE, RUET

8 of 27

Course Outcomes (COs):

8

Rizoan Toufiq, Assistant Prof, CSE, RUET

Develop the skills to gain a basic understanding of neural network theory, fuzzy logic and genetic algorithm.

Explore the functional components of neural network classifiers or controllers, and the functional components of fuzzy logic and genetic algorithm classifiers or controllers.

Creating and designing concepts of trainable neural network or fuzzy logic or genetic algorithm system to solve real world problem.

9 of 27

Grading:

Quizzes/Class Test: 20 Marks* (3 best out of 4 quizzes/class tests may be taken for awarding grade)
Homework’s/Attendance: 8 Marks*
Semester Exam: 72 Marks

* - We reserve the right to change the above grading scheme.

9

Rizoan Toufiq, Assistant Prof, CSE, RUET

10 of 27

Question Type:

10

Rizoan Toufiq, Assistant Prof, CSE, RUET

(Diagram 1.1, Bloom’s Taxonomy Revised)

The Cognitive Domain:

11 of 27

Question Type:

Remembering:

Recognizing or recalling knowledge from memory. Remembering is when memory is used to produce or retrieve definitions, facts, or lists, or to recite previously learned information.

Understanding:

Constructing meaning from different types of functions be they written or graphic messages or activities like interpreting, exemplifying, classifying, summarizing, inferring, comparing, or explaining.

11

Rizoan Toufiq, Assistant Prof, CSE, RUET

12 of 27

Question Type:

Applying:

Carrying out or using a procedure through executing, or implementing. Applying relates to or refers to situations where learned material is used through products like models, presentations, interviews or simulations.

Analyzing:

Breaking materials or concepts into parts, determining how the parts relate to one another or how they interrelate, or how the parts relate to an overall structure or purpose.

12

Rizoan Toufiq, Assistant Prof, CSE, RUET

13 of 27

Question Type:

Evaluating:

Making judgments based on criteria and standards through checking and critiquing. Critiques, recommendations, and reports are some of the products that can be created to demonstrate the processes of evaluation.
In the newer taxonomy, evaluating comes before creating as it is often a necessary part of the precursory behavior before one creates something.

13

Rizoan Toufiq, Assistant Prof, CSE, RUET

14 of 27

Question Type:

Creating:

Putting elements together to form a coherent or functional whole; reorganizing elements into a new pattern or structure through generating, planning, or producing.
Creating requires users to put parts together in a new way, or synthesize parts into something new and different creating a new form or product.
This process is the most difficult mental function in the new taxonomy.

14

Rizoan Toufiq, Assistant Prof, CSE, RUET

15 of 27

Question Type:

15

Rizoan Toufiq, Assistant Prof, CSE, RUET

The Cognitive Domain:

16 of 27

Question Type:

16

Rizoan Toufiq, Assistant Prof, CSE, RUET

The Cognitive Domain:

17 of 27

Question Type:

17

Rizoan Toufiq, Assistant Prof, CSE, RUET

Remembering	Define, Name, Describe, Outline, Identify, Recall, Label, Recite, List, Select, Match, State
Understanding	Convert Extend Defend Generalize Discriminate Infer Distinguish Paraphrase Estimate Predict Explain Summarize

18 of 27

Question Type:

18

Rizoan Toufiq, Assistant Prof, CSE, RUET

Applying	Change Organize Compute Prepare Demonstrate Relate Develop Solve Modify Transfer Operate Use

19 of 27

Question Type:

19

Rizoan Toufiq, Assistant Prof, CSE, RUET

Analyzing	Break down Infer Deduce Outline Diagram Point out Differentiate Relate Distinguish Separate out Illustrate Subdivide

20 of 27

Question Type:

20

Rizoan Toufiq, Assistant Prof, CSE, RUET

Evaluating	Appraise Judge Compare Justify Contrast Support Criticize Validate Defend

21 of 27

Question Type:

21

Rizoan Toufiq, Assistant Prof, CSE, RUET

Creating	Categorize Devise Compile Formulate Compose Predict Create Produce Design

22 of 27

Sessional

Course No: CSE 4204
Course title: Sessional based on CSE 4203
Prerequisite courses: None
Contact hours/week: 1.5
Credits: 0.75

Course Description

Sessional based on the theory of course CSE 4203.

*

Rizoan Toufiq, Assistant Prof, CSE, RUET

22

23 of 27

Sessional: Grading

Quiz Test: 20 Marks*
Homework's/Attendance: 8 Marks*
Board Viva: 25
Others: 47 Marks*

Lab Report (Individual)* - 20
Lab Performance (Individual)* - 27

*

Rizoan Toufiq, Assistant Prof, CSE, RUET

23

24 of 27

Sessional: Modules

*

Rizoan Toufiq, Assistant Prof, CSE, RUET

24

Module	Topics
01	Implementation of Nearest Neighbor classification algorithms with and without distorted pattern.
02	Design and implementation of single layer perceptron learning algorithm.
03	Design and implementation of Multi-layer Neural Networks algorithm (i.e., Back-propagation learning neural networks algorithm).
04	(1) Design and implementation of Kohonen Self-organizing Neural Networks algorithm. (2) Design and implementation of Hopfield Neural Networks algorithm.
05	Designing a Fuzzy Logic Controller (FLC) for controlling the environmental inputs to solve the real world problem.
06	Design and development of a Genetic Algorithm (GA) to search and optimize a specific problem.
07	Quiz/ Viva voce will be held during the Lab time.

25 of 27

Sessional: Modules

*

Rizoan Toufiq, Assistant Prof, CSE, RUET

25

Level	Description	Action Verbs Describing Learning Outcomes
Characteri-zation	All behavior displayed is consistent with one’s value system. Values are integrated into a pervasive philosophy that never allows expressions that are out of character with those values. Evaluation at this level involves the extent to which one has developed a consistent philosophy of life (e.g., exhibits respect for the worth and dignity of human beings in all situations).	Avoid�Display�Exhibit�Internalize�Manage�Require�Resist�Resolve�Revise
Organization	Commitment to a set of values. This level involves 1) forming a reason why one values certain things and not others, and 2) making appropriate choices between things that are and are not valued. One is expected to organize likes and preferences into a value system and then to decide which ones will be dominant.	Abstract Formulate�Balance Select�Compare Systemize�Decide Theorize�Define
Valuing	Display behavior consistent with a single belief or attitude in situations where one is neither forced or asked to comply. One is expected to demonstrate a preference or display a high degree of certainty and conviction.	Act Express�Argue Help�Convince Organize�Debate Prefer�Display
Responding	One is required to comply with given expectations by attending or reacting to certain stimuli. One is expected to obey, participate, or respond willingly when asked or directed to do something.	Applaud (show approval or praise by clapping.) Participate�Comply (act in accordance with a wish or command.) Play�Discuss Practice�Follow Volunteer�Obey
Receiving	One is expect to be aware of or to passively attend to certain stimuli or phenomena. Simply listening and being attentive are the expectations.	Attend Listen�Be aware Look�Control Notice�Discern Share�Hear

26 of 27

Website

*

Rizoan Toufiq, Assistant Prof, CSE, RUET

26

https://sites.google.com/site/t062000/nnfs17

27 of 27

End