Visual Analytics for Big Data
Faculteit  Science and Engineering 
Jaar  2017/18 
Vakcode  WMCS16000 
Vaknaam  Visual Analytics for Big Data 
Niveau(s)  master 
Voertaal  Engels 
Periode  semester I a 
ECTS  5 
Rooster  rooster.rug.nl 
Uitgebreide vaknaam  Visual Analytics for Big Data  
Leerdoelen  After successfully completing the course, students will be familiar with the theory and practice of visual analytics and its application to data science. They will have a good understanding of data representation issues (data types, storage schemes, performance vs scalability tradeoffs, and data interpolation and aggregation), data mining techniques (database queries, data clustering techniques), and information visualization techniques and tools that form the core of visual analytics (table lenses, sparklines, dimensionality reduction, timelines, tree/graph visualization algorithms, parallel coordinate plots, scatterplot matrices). They will be able to select the right mix of techniques and tools for data representation, processing, and visual exploration to construct, evaluate, and use visual analytics solutions. Finally, they will be able to apply such endtoend applications in problemsolving scenarios involving realworld datasets and questions.  
Omschrijving  This course discusses the challenges, opportunities, and solutions related to the visual analysis of big data collections. The course is structured along the classical pipeline of making sense from large and complex data collections using combinations of data mining, data analysis, and data visualization tools and techniques. The course starts presenting the context of big data, datacentric problemsolving, and hypothesis refinement, thereby defining the visual analytics sensemaking pipeline. The second module covers data representation (data types, data storage) with a focus on the main data types in visual analytics: tables, multivariate data, time series, trees, graphs, and text. The third module covers data processing (data mining, cleaning, querying, and aggregation), and related tools (SQLite, Spark, Hadoop, Hive). The subsequent three module cover the visual analysis of multivariate data, time series, and graphs/trees. The last two modules of the course are dedicated to the design and use of endtoend visual analytics applications based on realworld usecases.  
Uren per week  
Onderwijsvorm  Hoorcollege (LC), Practisch werk (PRC)  
Toetsvorm 
Presentatie (P), Schriftelijk tentamen (WE), Verslag (R)
(All three components (theory exam, report, end presentation) count equally towards the final grade.) 

Vaksoort  master  
Coördinator  prof. dr. A.C. Telea  
Docent(en)  prof. dr. A.C. Telea  
Entreevoorwaarden  Upon entering this course, students should 1. be familiar with a mainstream programming language (C, C++, Java, Python, C#, JavaScript). Basic proficiency is required, i.e. being familiar with multidimensional arrays, trees, lists, sorting algorithms, and graph traversal algorithms. 2. be familiar with basic computer graphics principles, techniques, and implementation tools (2D graphics pipeline, rendering, color models, polylines and polygons, affine transformations) 3. have a mathematical and statistics background including univariate and bivariate function analysis, derivatives, gradients, histograms, normal distributions, matrixvector products, inner/outer products, matrix inverses, and principal component analysis. 

Opmerkingen  
Opgenomen in 
