T LSANSST DERS KATALOG FORMU

(GRADUATE COURSE CATALOGUE FORM)

Dersin Ad Makine renmesi Course Name Machne Learnng Ders Seviyesi (Course Level) Y.L. (M.Sc.)

Yaryl Kredisi AKTS Kredisi (Semester) (Local Credits) (ECTS Credits) Gz 3 7.5 (Fall) Bilgisayar Mhendilii Lisansst Program (Computer Engineering) (Graduate Program) Dersin Tr (Course Type) Dersin erii (Course Description) 30-60 kelime aras Zorunlu (Compulsory) Dersin Dili (Course Language)

Kodu (Code) BLG527E

ngilizce (English)

Yapay renmeye giri; olaslk, istatistik ve lineer cebir hatrlatmas; eitmenli renme; Bayes renmesi; parametrik yntemler; ok znitelikli renme; boyut azaltma; kmeleme; parametrik olmayan yntemler; karar aalar; lineer snflandrma; karar destek makinalar; yapay sinir alar; hidden Markov modelleri; farkl renme yntemlerinin deerlendirme, karlatrmas; snflandrc birletirilmesi Introduction, review of probability, statistics and linear algebra; Supervised Learning; Bayesian Decision Theory; Parametric Methods ; Multivariate Methods; Dimensionality Reduction; Clustering; Nonparametric Methods; Decision Trees; Linear Dicrimination Support Vector Machines; Multilayer Perceptrons; Hidden Markov Models; Assessing and Comparing Classification Algorithms;Combining Multiple Learners Bu ders rencilere makine renmesinin temelleri, algoritmalar ve teknikleri hakknda bilgi verecek ve bu teknikleri kendi yazacaklar programlarla ve aratrma ile pratik ve teorik olarak kullanmalarn salayacaktr. This course will provide students an overview of issues, algorithms and techniques in machine learning. Students will also gain theoretical and practical experience through programming exercises and projects. 1. 2. 3. 4. Bir problemin zm iin yapay renme yntemi nerebilme Deiik yapay renme yntemlerinin performanslarn analiz edebilme Deiik yapay renme yntemlerinin sonularn birletirebilme Deiik yapay renme yntemlerininin teorik ve algoritmik olarak nasl altklarn anlayabilme

Dersin Amac (Course Objectives) Maddeler halinde 2-5 adet

Dersin renme ktlar (Course Learning Outcomes) Maddeler halinde 4-9 adet

1. 2. 3. 4.

Ability to propose a machine learning method for a specific problem Ability to analyze performance of different machine learning methods Ability to combine outputs of different machine learning methods Ability to understand the theoretical foundations and workings of different machine learning methods

Kaynaklar (References) En nemli 5 adedini belirtiniz devler ve Projeler (Homework & Projects)

[1] Alpaydin, E., 2004. Introduction to Machine Learning (Adaptive Computation and Machine Learning), The MIT Press. [2] Bishop, C., 2006. Pattern Recognition and Machine Learning, Springer. [3] Duda, R. O., Hart, P. E., and Stork, D. G., 2000. Pattern Classification, 2nd Edition, Wiley. Hw1:probablty, bayes learnng Hw2:parametric methods, multvarate methods, dmensonalty reducton Hw3:support vector machne, multlayer perceptron renilen yntemlerin hepsini ieren dnem projesi Hw1:olaslk, bayes renmesi Hw2:parametrik yntemler, ok boyutlu yntemler, boyut azaltma Hw3:support vector machne, yapay sinir a Course project using possbly all methods learned --Matlab ve Javada Uygulamalar

Laboratuar Uygulamalar (Laboratory Work) Bilgisayar Kullanm (Computer Use) Dier Uygulamalar (Other Activities) Baar Deerlendirme Sistemi (Assessment Criteria)

Matlab and/or Java Programmng --Faaliyetler (Activities) Yl i Snavlar (Midterm Exams) Ksa Snavlar (Quizzes) devler (Homework) Projeler (Projects) Dnem devi/Projesi (Term Paper/Project) Laboratuar Uygulamas (Laboratory Work) Dier Uygulamalar (Other Activities) Final Snav (Final Exam)

Adedi* (Quantity) 1

Deerlendirmedeki Katks, % (Effects on Grading, %) % 25 (25 %)

3 1

% 30 (30 %) % 20 (20 %)

% 25 (25 %)

*Yukarda Belirtilen Saylar Minimum Olup Yerine Getirilmesi Zorunludur.

DERS PLANI
Hafta 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Konular
Yapay renmeye giri; olaslk, istatistik ve lineer cebir hatrlatmas; eitmenli renme; Bayes renmesi; parametrik yntemler; ok znitelikli renme; boyut azaltma; kmeleme; parametrik olmayan yntemler; karar aalar; lineer snflandrma; karar destek makinalar; yapay sinir alar; hidden Markov modelleri; farkl renme yntemlerinin deerlendirme, karlatrmas; snflandrc birletirilmesi

Dersin ktlar 1,4 1,4 1,4 1,4 1,4 1,4 1,4 1,4 1,4 1,4 1,4 1,4 2 3

COURSE PLAN
Weeks 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Topics
Introduction, review of probability, statistics and linear algebra; Supervised Learning; Bayesian Decision Theory; Parametric Methods ; Multivariate Methods; Dimensionality Reduction; Clustering; Nonparametric Methods; Decision Trees; Linear Dicrimination Support Vector Machines; Multilayer Perceptrons; Hidden Markov Models; Assessing and Comparing Classification Algorithms; Combining Multiple Learners

Course Outcomes 1,4 1,4 1,4 1,4 1,4 1,4 1,4 1,4 1,4 1,4 1,4 1,4 2 3

Dersin Bilgisayar Mhendislii Yksek Lisans Programyla likisi

Programn mezuna kazandraca bilgi, beceri ve yetkinlikler (programa ait ktlar) i. ii. iii. iv. v. vi. vii. viii. ix. x. xi. xii. xiii. xiv. xv. xvi. Lisans dzeyi yeterliliklerine dayal olarak, Bilgisayar Mhendislii alannda bilgilerini uzmanlk dzeyinde gelitirebilme ve derinletirebilme (yeterli bilgi birikimi) (bilgi). Bilgisayar Mhendislii alannn ilikili olduu disiplinler aras etkileimi kavrayabilme (bilgi). Bilgisayar Mhendislii alannda edindii uzmanlk dzeyindeki kuramsal ve uygulamal bilgileri kullanabilme (beceri). Bilgisayar Mhendislii alannda edindii bilgileri farkl disiplin alanlarndan gelen bilgilerle btnletirerek yorumlayabilme ve yeni bilgiler oluturabilme (beceri). Bilgisayar Mhendislii alann ile ilgili karlalan sorunlar aratrma yntemlerini kullanarak zmleyebilme (beceri). Bilgisayar Mhendislii alan ile ilgili uzmanlk gerektiren bir almay bamsz olarak yrtebilme (Bamsz alabilme ve Sorumluluk Alabilme Yetkinlii). Bilgisayar Mhendislii alan ile ilgili uygulamalarda karlalan ve ngrlemeyen karmak sorunlarn zm iin yeni stratejik yaklamlar gelitirebilme ve sorumluluk alarak zm retebilme (Bamsz alabilme ve Sorumluluk Alabilme Yetkinlii). Bilgisayar Mhendislii alan ile ilgili sorunlarn zmlenmesini gerektiren ortamlarda liderlik yapabilme (Bamsz alabilme ve Sorumluluk Alabilme Yetkinlii). Bilgisayar Mhendislii alannda edindii uzmanlk dzeyindeki bilgi ve becerileri eletirel bir yaklamla deerlendirebilme ve renmesini ynlendirebilme (renme Yetkinlii). Bilgisayar Mhendislii alanndaki gncel gelimeleri ve kendi almalarn, nicel ve nitel veriler ile destekleyerek, alanndaki ve alan dndaki gruplara, yazl, szl ve grsel olarak sistemli biimde Trke ve/veya ngilizce olarak aktarabilme (letiim ve Sosyal Yetkinlik). Sosyal ilikileri ve bu ilikileri ynlendiren normlar eletirel bir bak as ile inceleyebilme, gelitirebilme ve gerektiinde deitirmek zere harekete geebilme (letiim ve Sosyal Yetkinlik). Bilgisayar Mhendislii alannn gerektirdii dzeyde bilgisayar yazlm ile birlikte biliim ve iletiim teknolojilerini ileri dzeyde kullanabilme (letiim ve Sosyal Yetkinlik). Bilgisayar Mhendislii alan ile ilgili verilerin toplanmas, yorumlanmas, uygulanmas ve duyurulmas aamalarnda toplumsal, bilimsel, kltrel ve etik deerleri gzeterek denetleyebilme ve bu deerleri retebilme (Alana zg Yetkinlik). Bilgisayar Mhendislii alan ile ilgili konularda strateji, politika ve uygulama planlar gelitirebilme ve elde edilen sonular, kalite sreleri erevesinde deerlendirebilme (Alana zg Yetkinlik). Bilgisayar Mhendislii alannda zmsedikleri bilgiyi, problem zme ve/veya uygulama becerilerini, disiplinleraras almalarda kullanabilme (Alana zg Yetkinlik). Tezli programlarda, kendi almalarn, Bilgisayar Mhendislii alanndaki uluslararas platformlarda, yazl, szl ve/veya grsel olarak aktarabilme (Alana zg yetkinlik). 1: Az, 2. Ksmi, 3. Tam Katk Seviyesi 1 2 3 X X

X X X

X X

X X

Relationship between the Course and Computer Engineering Graduate (MS) Curriculum
Level of Contribution 1 2 3 X X

Program Outcomes i. ii. iii. iv. v. vi. vii. viii. ix. x. xi. xii. xiii. xiv. xv. xvi. Developing and intensifying knowledge in Computer Engineering area, based upon the competency in the undergraduate level (sufficient knowledge) (knowledge). Grasping the inter-disciplinary interaction related to Computer Engineering area (knowledge). The ability to use the expert-level theoretical and practical knowledge acquired in Computer Engineering area (skill). Interpreting and forming new types of knowledge by combining the knowledge from Computer Engineering area and the knowledge from various other disciplines (skill). Solving the problems faced in Computer Engineering area by making use of the research methods (skill). The ability to carry out a specialistic study related to Computer Engineering area independently (Competence to work independently and take responsibility). Developing new strategic approaches to solve the unforeseen and complex problems arising in the practical processes of Computer Engineering area and coming up with solutions while taking responsibility (Competence to work independently and take responsibility). Fulfilling the leader role in the environments where solutions are sought for the problems related to Computer Engineering area (Competence to work independently and take responsibility) Assessing the specialistic knowledge and skill gained through the study with a critical view and directing ones own learning process (Learning Competence). Systematically transferring the current developments in Computer Engineering area and ones own work to other groups in and out of Computer Engineering area; in written, oral and visual forms in turkish and/or english (Communication and Social Competency). Ability to see and develop social relationships and the norms directing these relationships with a critical look and the ability to take action to change these when necessary. (Communication and Social Competency). Using the computer software together with the information and communication technologies efficiently and according to the needs of Computer Engineering area (Communication and Social Competency). Paying regard to social, scientific, cultural and ethical values while collecting, interpreting, practicing and announcing processes of Computer Engineering area related data and the ability to teach these values to others (Area Specific Competency). Developing strategy, policy and application plans concerning the subjects related to Computer Engineering area and the ability to evaluate the end results of these plans within the frame of quality processes (Area Specific Competency). Using the knowledge and the skills for problem solving and/or application (which are processed within the area) in inter-disciplinary studies (Area Specific Competency). In the programs with thesis, the ability to present ones own work within the international Computer Engineering environments orally, visually and in written forms (Area Specific Competency).

X X X

X X

X X

1: Little, 2. Partial, 3. Full

Dzenleyen (Prepared by) Zehra ataltepe

Tarih (Date)
15 Haziran 2011

mza (Signature)