İbn Haldun Üniversitesi Taş Mektep Kampüsü
Eğitimin toplam süresi 30 saattir.
Başvuru İçin Son Tarih
8 Temmuz - 12 Temmuz
09:30-12:30 & 14:00-17:00
800₺ (KDV Dahil)
- Nüfus cüzdanı fotokopisi,
- Ödeme yapıldığına dair banka dekontu.
İbn Haldun Üniversitesi
Sürekli Eğitim Uygulama ve Araştırma Merkezi
Bakırköy Taş Mektep (İtalyan Köşkü) Kampüsü
Yeni Mahalle Mektep Arkası Sokak İlköğretim No:40/888 Bakırköy/İstanbul
Ödeme ve başvurunuzu online kredi kartı ile veya EFT/havale yöntemiyle yapabilirsiniz.
Kredi karti ile online ödeme yapabilmek için öncelikle aşağıdaki bağlantıdan başvuru sistemine kayıt olmanız gerekmektedir. Devamında çıkacak olan ödeme ekranında açıklama alanına mutlaka isim, soyadı ve başvurulan eğitimin adı belirtilmelidir.
EFT veya banka havalesi yoluyla ödeme yapmak isteyen katılımcılarımız aşağıda belirtilen hesabımızı kullanarak kayıtlarını gerçekleştirebilirler. Katılımcılardan ricamız açıklama alanına isim, soyadı ve başvurulan eğitimin adı ve e-posta adreslerini mutlaka belirtilmeleridir. EFT/havale yöntemiyle yapılan ödemelerde, ödemenizi yaptıktan sonra bu bilgileri firstname.lastname@example.org adresine göndermeniz gerekmektedir.
Hesap ( Alıcı ) Adı: İbn Haldun Üniversitesi
Banka Adı: Kuveyt Türk Katılım Bankası A.Ş.
Şube Adı: İkitelli Şubesi (Şube Kodu 29)
Hesap No: 94341690-7
IBAN: TR46 0020 5000 0943 4169 0000 07
Hedefler: This 5-day course will provide a hands-on introduction to machine learning through social scientific applications. Although theoretical foundations will be presented, the focus will be on the empirical aspects. The course is a mix of theory and data analysis. Statistical software R will be used on Jupyter notebooks to analyse real datasets in order to illustrate the methods. Along with the statistical material, the course also aims to equip students with two additional computational skills: data management and data visualization. R packages dplyr and ggplot2 will be introduced and used for these purposes throughout the course. There are no strict pre-requisites but some familiarity with quantitative analysis and coding would be useful.
Öğretim Yöntemleri: The course is taught at an intermediate statistical level. That being said, it is pretty much self- contained in the sense that someone with some level of numeracy could start from scratch. Although the course covers the technical aspects of the models introduced, the emphasis will be on application, i.e., model selection, coding and interpretation of the findings. We are not statisticians, after all!
All sessions will be practical in the sense that even the theoretical discussions will be presented around research questions and data. The applications are mostly chosen from real social science research questions but examples from other disciplines like marketing, medicine and engineering are sometimes mentioned.
Statistical software R will be used throughout the course. R is an open source programming language and freely available online. We will use R on Jupyter Notebook, a web-based interactivecomputational environment which acts like a graphical user interface for programming languages. I will distribute a handout before the first lecture and talk more about these technicalities in the first session, so don’t worry!
You will be given data analysis homework assignments. They will not be marked but we will go over together the following day. Attendance is crucial. If you miss a session, it would be quite difficult to catch up.
I will make my best to make the course as smooth and accessible as possible, but, at the end of the day, it will be demanding. That being said, I am most certain that the “gains” are going to outmeasure the “pains” and, hopefully, you will join the small minority of social scientists who can understand and implement state of the art data analysis methods.
Okumalar ve Kaynaklar: There is no single textbook we will follow from beginning to the end, but the following book contains most of the material we will cover and shares a similar philosophy:
- James, Witten, Hastie and Tibshirani. An Introduction to Statistical Learning. Springer Press. 2013 (Corrected 2017) (Abbrev: ISL)
- Luckily, this book is freely available from the link below:
- The suggested readings are listed in the outline below. It would be very helpful if you could at least skim these before coming to the class and get an idea about what we will be - covering. I kept these readings at a minimum because the lecture notes will cover all the theoretical aspects you need to know. I might assign some other additional readings as the semester progresses.
- In terms of programming with R, apart from the codes I provide, Google is your friend. R has a very active online community and you can find solutions to all of your problems by simply searching the name of the function, the error message etc. The help files for the R packages are also very comprehensive and provides examples.
There are many other great books on the R and data science. You might find the following ones useful:
- A. Agresti. Foundations of Linear and Generalized Linear Models. John Wiley & Sons, 2015
- W. Chang. R Graphics Cookbook. O'Reilly Media, Inc., 2012.
- M.J. Crawley. The R Book. John Wiley & Sons, 2012.
- J. Friedman, T. Hastie, and R. Tibshirani. The Elements of Statistical Learning. Springer, 2001.
- D. Gutierrez. Machine Learning and Data Science: An Introduction to Statistical Learning Methods with R. Technics Publications, 2015
- N. Matloff. The Art of R Programming: A tour of statistical software design. No Starch Press, 2011.
- M. McGrath. R for Data Analysis in Easy Steps. In Easy Steps Limited, 2018. H. Wickham. ggplot2: Elegant Graphics for Data Analysis. Springer, 2016
- Principles of Quantitative Analysis and Machine Learning Introduction to R with Jupyter Notebooks
- Data: Input/Output, Types, Manipulation and Visualization
- Regression: Multiple and Polynomial
- Classification 1: Logistic Regression
- Classification 2: K-Nearest Neighbours and Tree-Based Methods
- Clustering: K-Means and Hierarchical Clustering
- Dimensionality Reduction: Principal Component Analysis and Linear Discriminant Analysis
- Natural Language Processing: Text as Data and Sentiment Analysis on Twitter Data
- Cross Validation and Model Selection
- Wrap-up and Open Discussion