Symbolic data analysis (SDA: Diday (1987); Bock and Diday (2000); Billard and Diday (2003); Noirhomme-Fraiture and Brito (2011)) has gained popularity over the past few years because of its potential for handling data with hierarchical or dependent nature and for handling big data sets. The major difference between symbolic and conventional data is that cells (rows by columns) in a data table might contain an interval, a histogram, or even a distribution. According to the type of symbolic data used, a statistical procedure for conventional data analysis has to be adapted to symbolic data analysis accordingly. In this we mean to include descriptive statistics, graphics and visualization, exploratory data analysis, statistical models, and inferences.

Recent development of SDA techniques ranges from theoretical foundations to all sorts of multivariate analyses (regression, PCA, discriminant analysis, clustering, MDS, exploratory data analysis and visualization) to real world applications (social surveys, biomedical experiments, engineering). With the potential for handling data with hierarchical or dependent nature SDA is ready for facing the new challenge of BIG DATA.

The workshop presents itself as an excellent opportunity for crossover of ideas between the community members and to open the way to future cooperation between them.

This workshop invites oral presentations that contain recent developments, from a theoretical or methodological point of view, practical applications and new software tools. The workshop will also include sessions for open discussions about the recent achievements and the open challenges in the field.