Data Analysis and Machine learning

Course goals

In this this course, you will learn to:

• know, explain, and apply data retrieval from existing relational and nonrelational databases, including text, using queries build from primitives such as select, subset, and join both directly in, e.g., SQL and Python Pandas.
• know, explain, and apply common data clean-up procedures, including missing data and the appropriate imputation methods and feature selection.
• know, explain, and apply methodology to properly set-up data analysis experiments, such as train, validate, and test and the bias/variance trade-off.
• know, explain, and apply supervised machine learning algorithms, both for classification and regression purposes as well as their related quality measures, such as AUC and the confusion matrix.
• know, explain, and apply non-supervised learning algorithms, such as clustering and other techniques that result in lower-dimensional data representations.
• be able to choose between the different techniques learned in the course and be able to explain why the chosen technique fits both the data and the research question best

Assessment
Your final grade in the course consists of the following grading components:

• biweekly assignments (20% of the final mark). Every two weeks, there is a group assignment. Each assignment is graded and worth 5% of the final grade.
• midterm exam (40%): Halfway through the course, there is a midterm exam. The content of this exam pertains to the first half of the course.
• final exam (40%): At the end of the course, there is a final exam. The content of this exam emphasizes non-exclusively the teaching material of the second half of the course.”

To pass the course, the weighted final grade of all grading components should be greater than or equal to 5.5.

Retake

• if you obtain a final grade between 4.0 and 5.4, you are eligible for a retake
• you can only retake one of the two exams. By default, this will be the one with the lowest grade. If you obtained the same grade for both, you will redo the final exam.
• the grade attained in the retake will replace the grade from the selected exam.

Prerequisites
This course is for students in the master Applied Data Science only.

Content

Data do not fall from heaven, but are created, manipulated, transformed, and cleaned - in any data analysis, therefore, the treatment of the data itself is just as important as the modeling techniques applied to them.
In this course, you will get acquainted with and implement a variety of techniques to go from raw data to analyses, visualizations and insights for science and business applications. This is an overview course designed to give you the tools and skills to use and evaluate data science methods.

The course consists of two parts, data wrangling and data analysis, which are intertwined.

Course form
Each week there are lectures that present the theories and give a general overview of the systems that are available.
Then laboratory exercises and the tutorial sessions give a hands-on experience where the students can practice the theory on real-world applications. These laboratories and tutorial sessions are performed with the assistance of the teaching team. The practical work done in these labs is drawn from real life situations that allow the students to experience how to solve data science problems.

The contents of the course are available on https://infomdwr.nl/.

Literature
Tentative (can be changed during the course) core reading materials:

• James et al, "Introduction to Statistical Learning" http://www-bcf.usc.edu/~gareth/ISL/
• Grolund & Wickham "R for Data Science" https://r4ds.had.co.nz/
• Abraham Silberschatz, Henry F. Korth, S. Sudarshan "Database System Concepts"
• Jiawei Han, Micheline Kamber, Jian Pei "Data Mining: Concepts and Techniques"
• Jure Leskovec, Anand Rajaraman, Jeff Ullman, "Mining Massive Datasets" http://www.mmds.org
• Jurafsky, D., Martin, J.H., "Speech and language processing". third edition. Online chapters: https://web.stanford.edu/~jurafsky/slp3/
• other freely available literature