CMPE535 - Knowledge Engineering

Instructor Dr. Mehmet Bodur 

Announcements

25.6.2019:  Table of statistics updated:

verageCodeCategoryMaximumMinimumAverage
101QZ
99
50
83,17
202HW1
100
0
90,63
303HW2
100
70
93,96
404HW3
100
70
95
505MT
98
52
79,54
606FIN
93
58
81,54
707ATT
100
100
100

MT is highest of MT1 and MT2. 
Average of MT2  is 1% less than filtered average of MT1 for students who 
entered also MT2, indicating clearly MT2 was fair exam.

23.6.2019 HW and Final grades are on portal.
Some HWs had 70 for plagiarism in conclusion ! with reference,
and one HW is 0 for plagiarism in conclusion ! without references.

18.6.2019 Correction for 5.th question of Sample Questions on Fuzzy:  

answer shall be  Monotonicity of t (a, b) is tested by all a1 ≤ a2 and

b1 ≤ b2 to satisfy t(a1, b1 ) ≤ t(a2 , b2)

17.6.2019 Sample Questions on Fuzzy with Corrected Q10 and Q13 (PDF)

16.6.2019 Sample Questions on Artificial Neural Nets (PDF)

13.6.2019 Sample Questions on Fuzzy Knowledge Base Systems (PDF)

29.5.2019 MT2 Exam on 19.6.2019 Wed, 12:30 /  Final Exam on 20.6.2019 Thu. 12:30.

23.5.2019  TextBook for Fuzzy part (PDF, 3MB)

21.5.2019  HW3ANFIS in R modeling IRIS data (PDF) (Rmd-Zip 3MBdue to 30 May

21.5.2019  Slides for Artificial Neural Net Systems (PDF

21.5.2019  Slides for Fuzzy Knowledge Base Systems (PDF

30.4.2019  Corrections for Ch2 Sample Questions (PDF

25.4.2019 - Sample Questions ID3 Information Loss  Table (PDF)

25.4.2019 - Sample Questions on Ch-2 part 2 and 3 for MT (PDF)

20.4.2019- Slides fourth part (PDF)

20.4.2019- Sample Questions on Ch-1-part-1 for MT (PDF) 

9.4.2019- Midterm Date 30.Apr.2019 (Tue) 

29.3.2019- Sample Questions on Ch-1 for Quiz-1 (PDF)

27.3.2019- Quiz-1 date changed to 2 Apr. 2019. 

26.3.2019- Quiz-1 on Ch1 on 3 Apr. 2019 last 25 minutes of lecture hour (slides- first part).

26.3.2019- Home Work 2, CLIPS (PDF) (due date Apr.10)

26.3.2019- Slides first part (PDF), second part (PDF), third part (PDF)

12.3.2019- Non-registered Students: According to the Dean's Office in general, students who do not have temporary registration at the end of the period are not allowed to follow the lectures. Students with provisional registration may follow, and take examinations. Warning by Dean says "For students who have financial problems are obliged to make their payments (the first installment of the past debt and / or the new period) at a later date (except for those who did not fulfill their obligation to pay within the relevant period during the 2018-19 Fall semester). A temporary registration application will be taken in the Services Office on 22 February 2019. Students who do not perform normal or temporary enrollment and attend the courses indifferently will not be accepted for any courses related to their courses during the semester or at the end of the semester and the courses and lecture activities will not be accepted".

11.3.2019- Home Work 1 Data Sets (PDF). Due May 26

Course Description

An overview of AI, Knowledge-based systems-a survey; Knowledge Engineering concepts;  Human Problem Solving; Human Information Processing System; Cognition Models; Knowledge Acquisition; Knowledge Representation; Production Rules; Inference, Forward Chaining, Backward Chaining, Mixed Chaining; Uncertainty, Certainty Factors, Bayesian, Fuzzy set based and Dempster-Shafer methods; Automated Knowledge Acquisition, Machine Learning Approaches in Expert Systems, Rule and Decision-Tree Induction; Connectionist Expert Systems; Expert System Building Tools, Development languages, Shells, Environments; Expert system design using rule-based shells; Expert system development life-cycle; Blackboard architectures; Truth Maintenance Systems.

AI ve bilgi tabanlı sistemlere genel bakış; Bilgi Mühendisliği kavramları; İnsanlarda Problem Çözme ve Bilgi İşleme Sistemleri; Algılama Modelleri; Bilgi toplama; Bilginin temsili; Üretim Kuralları; Çıkarım, İleri Zincirleme, Geriye Zincirleme, Karışık Zincirleme; Belirsizlik, Kesinlik Faktörleri, Bayes, Bulanık kümeler ve Dempster-Shafer yöntemleri; Otomatik Bilgi Toplama, Uzman Sistemlerde Makinesel Öğrenme Yaklaşımları, Kural, ve Karar-Ağacı ile İndüksiyon; Bağlantısal Uzman-Sistemler; Uzman-Sistem Oluşturma Araçları ve Geliştirme dilleri, Kabuklar, Ortamlar; Kural-tabanlı kabuklar kullanarak uzman-sistem tasarımı; Uzman-sistem geliştirme yaşam-döngüsü; Yazı tahtası mimarileri; Doğruluk-Bakım Sistemleri.

Text Books 
1- Nikolai K. Kasabov,  Foundations of Neural Networks, Fuzzy Systems, and Knowledge Engineering,  Oct 1996  ISBN 0-262-11212-4,  550 pp. 282 illus. MIT Press.
2- Simon Kendal  Malcolm Creen, An Introduction to Knowledge Engineering. Springer-Verlag London Limited 2007.
3 -  Hans-Jürgen Zimmermann, Fuzzy Set Theoryand Its Applications, Ed.4, Springer Science+Business Media, LLC. 1996-2001, ISBN 978-94-010-3870-6, ISBN 978-94-010-0646-0 , 

Schedule 
Tuesday 14:30 and 15:30 @ CMPE036  &  Wednesday 13:30 @ CMPE035

Attendance and Make-Up Exam Policy:
Minimum 70% Attendance is credited by 3 points, 85% and over is credited by 5 points.
There is no make-up exam for Quizzes.  Students missing the Mid-term or Final exam must submit a legitimate excuse within 3 working days after the exam date in order to qualify for a make-up. Only one makeup exam will be given for one of the missed exams (midterm or final) at the end of the semester that will cover all the topics of the course.
Any attempt to cheating in exams and quizzes will start disciplinary action and result in fail (F).
Any attempt to plagiarise in homework and/or projects will set grade of all homeworks to zero, together with disciplinary action.

Grading Policy
Any trial of cheating will cause a discipline reaction and a fail in the course

Quizzes (Minimum Four)
20%

Attendance  
5%
Midterm Examination 
25%
Homework and Projects
20%
Final Exam
30%

Course Outline

Week, 
Topic
1
An overview of AI, Knowledge-based systems - a survey;  Knowledge Engineering concepts;

Human Problem Solving; Human Information Processing System .

Major issues in KE. Typical KE problems and tools .

KE and Symbolic Artifical Intelligence, Data, Information, Knowledge. Data Analysis,
5
Data and Knowledge Representation. Methods for Symbolic Manipulation and Inference. Forward Chaining, Backward Chaining, Mixed Chaining; Propositional Logic and PROLOG. Production Rules, Expert Systems; Architecture and Design, Knowledge Acquisition
6
Uncertainty, Certainty Factors, Bayesian, Fuzzy set based and Dempster-Shafer methods; Fuzzy Sets and Systems: Extension principle; Fuzzy Proposition and Fuzzy Logic; Fuzzy Rules and Inference; Fuzzy Databases; Clustering-Based Methods for Fuzzy Rule Extraction, Fuzzy Prediction
7
Midterm Exam

Neural Networks: Biological and Artifical Neurons; Supervised Learning; Recurrent Networks;
9
Unsupervised Learning; Kohonen Maps; Neural Networks as Associative Memories
10
Automated Knowledge Acquisition, Machine Learning Approaches in Expert Systems, Rule and Decision-Tree Induction;
11
Connectionist Expert Systems;
12
Expert System Building Tools, Development languages, Shells, Environments;
13
Expert system design using rule-based shells;
14
Expert system development life-cycle; Blackboard architectures; Truth Maintenance Systems.