Examination arrangement

Course content

Course content:

Research shows that humans are much better at analyzing visual information, in comparison to plain numbers. Effective visualization can help to better understand the data, underlying patterns, trends and connections. Visualization is one important part of Data Science.

However, data often comes in a raw form, is usually noisy, inaccurate, incomplete and sometimes contradicting. Therefore, it is important to familiarize with the preparation phase – to clean and preprocess data, to extract information out of it. Visualization is a central part in the whole data analysis cycle: from visual analytics during the cleansing process to rich and interactive visualizations.

The course consists of two main parts - data processing and data visualization, with the following topics:

1. Data processing:

- Typical problems of raw data.
- Data cleansing methods.
- Different data formats, conversion and aggregation.
- Simple visualization and statistical tools for data quality inspection.
- Introduction to analog and digital signal processing.
- Signal sampling and synchronization.
- Fourier analysis and Fast Fourier transform.
- Basic filter design.

2. Data visualization:
- Human perception.
- User interface design principles.
- Charts and their applicability for different data types.
- Interactive visualizations.
- Modern visualization tools.
- Visualization of specific data sets (multi-dimensional data, geo-spatial data).

Learning outcome

Knowledge:

- Recall the basic principles of human perception.
- Recall fundamental data and signal processing methods.
- Understand how to clean and validate data.
- Understand how data visualization guidelines depend on human perception.

Skills:
- Apply data and signal processing and visualization tools to clean and visualize a data set.
- Experiment with different visualization techniques.

General competence:
- Analyze data using interactive exploration and visual analytics.
- Evaluate and choose the best visualization tools and techniques for a given data set.
- Evaluate a data visualization created by others.
- Create a story using a visualization of a dataset.

Learning methods and activities

The course will include lectures and self-study material on the course topics, and individual assignments on data processing and visualization. The students will work with several data sets, process and visualize them.

Grading will be completely based on the assignment reports.

The assignments will be distributed over the whole semester. The specific number and schedule of the assignments will be given at the start of the course. Students must deliver all assignments on time to pass the course.

The assignments may include peer reviews.

Compulsory assignments

• Mandatory assignment

Further on evaluation

Evaluation:

The grade will be determined as a weighted average of all course assignment grades.


Grading:

Grading A-F. Grade F is a fail

Specific conditions

Course materials

Books:
1.Tamara Munzner, Visualization Analysis & Design. CRC Press.
2. Jeff Johnson. Designing with the Mind in Mind, 2nd Edition. Morgan Kaufmann.
3. Steven W. Smith. The Scientist & Engineer's Guide to Digital Signal Processing. California Technical Pub.
Online courses and tutorials:
1. John Hopkins University. Regression models.
https://www.coursera.org/learn/regression-models
2. D3.js tutorials.
https://github.com/d3/d3/wiki/Tutorials