Data preprocessing and visualisation, 7.5 Credits

Contents

The objective of Data Science is to enable society, companies and citizens to understand and use the ever-increasing amount of collected data in ways that make it possible to detect potential problems or improvements to the current state of affairs. Data Science should also empower humans to estimate and understand the potential result of different actions. There's a saying about "lies, damned lies, and statistics", which expresses the fact that data-based statistics can be presented in very convincing ways even when the conclusions are false. This course attempts to teach how to detect such false information and ensure more ethical use of Data Science

One example of practical use of Data Science is analyzing and presenting epidemic-related data and statistics in correct and human-understandable ways so that decisions and actions can be taken based on rational information. Data Science methods are also used for estimating effects of actions for reducing global warming, dimensioning road networks, choosing where to install new shopping centers or restaurants, optimizing the energy usage of buildings, …. To put it shortly, Data Science is one of the most crucial domains for deciding how our current and future society is to be built. More and more companies are also coming to realize the importance of Data Science. Regardless of industry or size, organizations that wish to remain competitive in the age of big data need to efficiently develop and implement Data Science capabilities or risk being left behind

Module 1, theory, 4.0 credits.
This course on data preprocessing and visualization provides an introduction to the domain of Data Science. The students will learn how to import, manipulate and preprocess data coming from various real-world data sources with the objective to present it in ways that allow gaining insight into the underlying systems or phenomena. Preprocessing of data may produce improved insight into the meaning of data by statistical measurements, presented as numerical tables that summarize the data in various ways. However, in most cases, humans tend to understand visual presentations of data better than purely numerical presentations. The course will teach how to use basic data visualizations such a point and line plots, bar charts, histograms, boxplots and violin plots. 3D visualization techniques will be taught, as well as how to use maps and images for data visualization.

Various data analysis and machine learning methods will be used but the underlying theory is beyond the scope of this course. The intention is to make the students proficient with how those methods can be applied in real-world settings encountered in industry and society in general. This is why lectures are accompanied by exercises where students practice applying some of the methods treated during lectures.

The course mainly uses the R programming language, so students will learn the basics of R. Also included is an introduction to how data preprocessing and visualization methods can be used in the Python programming language.

Topics covered are:

• Introduction to the R programming language and tools
• Introduction to data processing and visualization in the Python programming language
• Import and export of data from text files, data bases and other sources
• Data visualization in R, in 2D and 3D
• Map visualizations
• Displaying and working with images in R
• Introduction to other useful data preprocessing and visualization packages
• Linear regression, BLUE, RMSE, shrinkage methods (Lasso, ridge regression)
• Linear classification (logistic regression, LDA)
• Principal Components Analysis (PCA) for identifying linear correlations between variables
• K-means clustering
• Nonlinear or nonparametric methods (e.g., k-NN)
• Preparation of data for machine learning
• Basic notions of Explainable Artificial Intelligence (XAI) 

Module 2, proficiency training, 3.5 credits.
Module 2 consists in a practical project that requires the combined use of methods learned in Module 1. Project topics and data sets will be provided by the course personnel, but student-proposed topics are encouraged. The project is performed in groups of 1-4 students. Each group presents their progress, plans and open questions to course personnel and fellow students in two "mentoring sessions" and in one final presentation session. The purpose of mentoring sessions is to provide constructive feedback and guidance to the students in their learning project. Mentoring session do NOT directly influence the grading of this Module.

Expected learning outcomes

Knowledge and understanding
After completing the course, the student should be able to:

• (FSR 1) understand what is meant by Data Science as a concept: where and when Data Science is needed, what types of problems Data Science can solve and what main methodologies and tools of Data Science are,
• (FSR 2) understand the meaning of various data-based measurements and visualizations commonly used in society, and know how to read and interpret them.

Competence and skills
After completing the course, the student should be able to:

• (FSR 3) understand data structures in the R programming language and have basic notions of data manipulation and programming in R,
• (FSR 4) perform manual as well as automated pre-processing of data (cleaning, normalization, centering, scaling, …),
• (FSR 5) extract and understand statistical indicators from data, as well as how to detect and eliminate missing values,
• (FSR 6) perform regression analysis and clustering of data,
• (FSR 7) visualize data and results of analyses using line plots, scatter plots, bar plots. maps etc., both in 2D and 3D.

Judgement and approach
After completing the course, the student should be able to:

• (FSR 8) assess the correctness and significance of data-based measurements and visualizations encountered in various media.

Required Knowledge

At least 7.5 ECTS mathematical statistics.

Form of instruction

The course consists of lectures, practical exercises performed individually, and a project performed in groups of up to four students. In addition to scheduled activities the course also requires individual work with the material

Examination modes

The assessment of Module 1 (ELO 1-7) is done through a written Learning Diary, which includes written lab reports. The grades given in this module are Fail (U), Pass (G) or Pass with distinction (VG).

The assessment of Module 2 (ELO 3-8) is done through a written project report. The grades given in this module are Fail (U), Pass (G) or Pass with distinction (VG).

On the whole course, one of the grades Fail (U), Pass (G) or Pass with distinction (VG) is given. The grade given on the course is Pass with distinction (VG) only if both Modules have the grade Pass with distinction (VG).

Adapted examination
The examiner can decide to deviate from the specified forms of examination. Individual adaptation of the examination shall be considered based on the needs of the student. The examination is adapted within the constraints of the expected learning outcomes. A student that needs adapted examination shall no later than 10 days before the examination request adaptation from the Department of Computing Science. The examiner makes a decision of adapted examination and the student is notified.

Other regulations

If the syllabus has expired or the course has been discontinued, a student who at some point registered for the course is guaranteed at least three examinations (including the regular examination) according to this syllabus for a maximum period of two years from the syllabus expiring or the course being discontinued.