Selected topics in Data Science and Machine Learning (2019)

Course organization

• place: Pasteura 7, room 404
• time: Friday, 14:00-16:30
• info and materials: this page
• format: lectures and classes

Lectures

• lecture 1 (05.04.2019)
• lecture 2 (26.04.2019)
• lecture 3 (10.05.2019)
• lecture 4 (24.05.2019)
• lecture 5 (31.05.2019)
• lecture 6 (07.06.2019)
• lecture 7 (24.06.2019)
• lecture 8 (13.09.2019)
• lecture 9 (11.10.2019)
• lecture 10 (15.10.2019)

Lectures

• List of problems up to lecture 7

Additional materials

Articles and demos

• TensorFlow Demo
• ROC curve applet
• Some Studies in Machine Learning using the Game of Checkers - Arthur Samuel (1959)
• AI, Deep Learning, and Machine Learning: A Primer - Frank Chen
• Building high-level features using large scale unsupervised learning - Quoc V. Le et al. (2012)
• Graphs in Statistical Analysis -F. J. Anscombe (1973). See also the local copy here
• On Discriminative vs Generative classifiers - a comparison of logistic regression and naive Bayes -A. Ng and M. I. Jordan See also the local copy here
• The Optimality of Naive Bayes - H. Zhang
• An introduction to ROC analysis - Tom Fawcett (2005)
• The relationship between Precision-Recall and ROC curves - J. Davis and M. Goadrich
• The Unreasonable Effectiveness of Data - A. Halevy, P. Norvig, and F. Pereira

Programming

In general programming in Python is easy and one can start very quickly. Of course some previous experience in programming is very helpful. For our course except from the standard Python, it would be useful to learn a bit about numpy and matplotlib.

• Official Python community website
• A lot of tutorials
• Dive into Python. I would recommend this book for people who feel that they have some experience in programming.
• Numpy introduction
• Matplotlib
• scikit-learn
• Python style guide

Datasets

Funny and tricky example based on the F. J. Anscombe's article (see above). This is a nice demonstration that the purely numerical analysis might be also misleading and that the data visualization plays an important role.

• data1.csv

"Classical" flower classification problem:

• iris_data.csv

Some test data set

• trainingColor.csv

Linear regression examples (see more here)

• x03.txt and x03.csv
• x08.txt and x08.csv
• x26.txt and x26.csv