Lista de Disciplinas | CMP273

CMP273 –  Visual Analytics for Data Science

Responsável: João Luiz Dihl Comba

Pré-requisitos: –

Carga Horária: 60 hs

Créditos: 4

Matrícula de Graduandos: A matrícula será feita como aluno especial

SÚMULA: Visual analytics techniques are playing a key role in the data science analysis and discovery process. In this course we focus on state-of-the-art techniques that employ visual analytics techniques in a variety of data science applications, including text analysis, cluster analysis, machine learning, big data, high-dimensional data, urban data, sports, healthcare, medicine, and immersive applications, among others.

OBJETIVOS: The goal of this course is to review visual analytics techniques for data science analysis. The course focus lies in the presentation of current work in a multitude of data science fields, which can be useful for students pursuing varying research fields. The students are expected to conduct a hands-on final project on a topic of choice related to their research interests.

CONTEÚDO PROGRAMÁTICO:

• Introduction: Visual Analytics in Data Science
• Visual Analytics and Data Science Basics
• Tools for Visualization and Data Analysis
• Examples of Visual Analytics projects
• Study of Visual Analytics in Data Science papers: Clustering Analytics, Text Visual Analytics, Machine Learning Visual Analytics, Healthcare and Biology Visual Analytics, Social Media Visual Analytics, Urban Data Visual Analytics, Sports Visual Analytics, among others.
• Final Project Development

TÉCNICAS DE ENSINO: The course is centered on the presentation of basic concepts of visual analytics and data science, presentation of visual analytics projects, and discussion of current papers describing visual analytics techniques for data science.

CRITÉRIOS DE AVALIAÇÃO:  The course requires that the students actively participate in the discussions in class, present selected, and develop and present a final project describing visual analytics techniques to a data science application.

The final grade is computed as follows:

• In-class participation (10%)
• Paper presentations (40%)
• Final Project (50%)

BIBLIOGRAFIA: The bibliography is composed of papers and textbooks:

• Papers: State-of-the-art papers in the field of visual analytics for data science, from conferences such as IEEE VISUALIZATION, EUROVIS, KDD, VLDB.
• Textbooks:
• Visual Analytics for Data Scientists. Andrienko, N., Andrienko, G., Fuchs, G., Slingsby, A., Turkay, C., Wrobel, S. Springer 2020.
• The Data Science Design Manual. Steven Skiena. Springer 2017.