Course information

ISSS608 Visual Analytics and Applications

About

Weekly Session

DataViz Makeover

Assignment

Visual Analytics Project

Resources

Synopsis

In this competitive global environment, the ability to explore visual representation of business data interactively and to detect meaningful patterns, trends and exceptions from these data are increasingly becoming an important skill for data analysts and business practitioners. Drawing from research and practice on Data Visualisation, Human-Computer Interaction, Data Analytics, Data Mining and Usability Engineering, this course aims to share with you how visual analytics techniques can be used to interact with data from various sources and formats, explore relationship, detect the expected and discover the unexpected without having to deal with complex statistical formulas and programming.

The goals of this course are:

• To share with you the principles, best practices and methods of visual analytics
• To provide you hands-on experiences in using commercial-off-the-shelf visual analytics software and programming tools to design visual analytics applications

Learning Objectives

Upon successful completion of the course, students will be able to:

• Understand the basic concepts, theories and methodologies of visual analytics.
• Analyse data using appropriate visual thinking and visual analytics techniques.
• Present data using appropriate visual communication and graphical methods.
• Design and implement cutting-edge visual analytics system for supporting decision making.

Competencies

• Explaining the concepts and principles of Visual Analytics.
• Describing the differences between Visual Analytics, Data Visualisation, Statistical Graphs and Infographics.
• Explaining the basic concept of visual variables and applying these concepts and best practice in designing data-driven static graphs.
• Explaining interactive techniques and best practice, and applying these techniques in designing interactive data visualisation.
• Understanding the data characteristics of numerical data and building data visualisation by using appropriate visually driven univariate and bivariate data analytics methods.
• Understanding the characteristics of multivariate data and building data visualisation by using appropriate multivariate visualisation methods.
• Understanding the characteristics of time-series data and building data visualisation by using appropriate time-series visualisation methods.
• Understanding the characteristics of geographical data and building data visualisation by using appropriate geovisualisation methods.
• Understanding the characteristics of network data and building data visualisation by using appropriate network graph visualisation methods.
• Explain the concepts and principles of Information Dashboard.
• Building analytical dashboard by using Commercial off-the-shelf (COTS) software.
• Designing visual analytics application programmatically by using free and open source software and packages.

Basic Modules

This course comprises ten integrated components as shown below:

Prerequisites

There are no prerequisites for the class. However, students taking this course must be willing to learn R programming framework. For students who are new to R, you are encouraged to consult the following resources prior to the lesson starts:

• Pedro J. Aphalo (2010-2017) Learn R .. as you learnt your mother tongue, Leanpub. (https://leanpub.com/learnr).
• R语言忍者秘笈
• R for Data Science

Course Assessment

The assessment of this course consists of four major components, namely:

• class participation,
• in-class hands-on exercise,
• DataViz makeover
• individual assignments,
• visual analytics project.

There will be no mid-term test or final examination for this course.

Class Participation

A strict requirement for each class meeting is to complete the assigned readings and to try out the hands-on exercises before coming to class. Readings will be provided from the textbook on technical information and from provided documents and articles on business applications of Visual Analytics. Students are required to review the recommended readings and class exercises before coming to class. Without preparation, the learning and discussions would not be as meaningful. Student sharing of insights from readings and hands-on exercises of assigned materials in class participation will form a large part of the learning in this course.

DataViz Makeover

Each week, I will post one or two data visualisation and you are required to critic, suggest ways for improvement and rework the data visualisation. Maybe you retell the story more effectively, or find a new story in the data. I am curious to see the different approaches you all take.

The purpose of the makeover is to improve on the original visualisation. Focus on what works, what doesn’t work, why those things don’t work, and how you made it better. You should try stick to the fields in the data set provided and improve upon the original visualisation. However, if supplementing the data helps you tell a better story, go for it!

You are required to upload the weekly makeover write-up onto the dropbox of e-Learn (i.e. LMS) and the data visualisation onto Tableau Public by Wednesday before mid-night 11.59pm. The write-up must be in MS Word format. You are also required to provide the URL link to the Tableau Public on eLearn (i.e. DataViz Makeover 1).

Individual Assignments

There is one assignment that is due throughout the term. Assignment due is to be uploaded into the course wiki strictly before the official due dates. Late work will be severely penalised. Students must check and confirm on Wiki the assignment due dates.

The assignment will be graded on a scale from 0 to 100. Scores of 70 and 80 are given when the assignment is essentially done completely and correctly. Scores 80 and 90 are reserved for complete and correct homework where extra initiative or innovation clearly sets the completed work above the simple, perfunctory and satisfactory completion of the mini-assignment.

Visual Analytics Project

The purpose of the project is to provide students first hand experience on collecting, processing and analysing large business data using real world data. A project may involve developing new methods or implementing visual analytics system to support analytic tasks in specific domains. Alternatively, a project may be in the form of application development by integrating analytical tools within a visual analytics environment. Students are encouraged to focus on research topics that are relevant to their field of study. It should address a concrete visual analytics problem and should propose a novel and creative solution.

For more details please refer to Visual Analytics Project page

Grading Summary

The grading distribution of this course is as follows:

• Class Participation 10%
• DataViz Makeover 35%
• Assignment 25%
• Visual Analytics Project 30%
  • Wiki 15%
  • Poster 15%
  • Workshop 20%
  • Practice research paper 25%
  • Artifact 25%

There will be no mid-term test or final examination for this course.