Nested Intellectual Data Grouping and Clusterization for the Interactive Visual Explorer

Maria Grigorieva,Igal Milman,Mikhail Titov,Timofei Galkin,Gh Adam,M Hnatič,J Buša

doi:10.1051/epjconf/202022603011

Abstract

The Interactive Visual Explorer (InVEx) application is designed as a visual analytics tool for Big Data analysis. Visual analytics is an integral approach to data analysis, combining methods of intellectual data analysis with advanced interactive visualization. One of the main objectives of InVExis to process large data samples by decreasing their level of detail (LoD).The proposed approach includes clustering as well as flexible grouping by different parameters, providing the exploration of data from the lowest to the highest level of details. The results of grouping and clusterization arevisualized using interactive 3D scene and parallel coordinates, allowing the user to gain insight into data, to explore hidden correlations and trends of parameters.

Highlights

The Interactive Visual Explorer (InVEx) is developed as a generic interactive visual analytics tool for the analysis and exploration of big volumes of multidimensional data [1]
InVEx is based on the combined usage of intellectual data analysis methods and advanced interactive visualization techniques
That’s where visual analytics may help by combining intellectual data analysis and interactive visualization

Summary

Introduction

The Interactive Visual Explorer (InVEx) is developed as a generic interactive visual analytics tool for the analysis and exploration of big volumes of multidimensional data [1]. InVEx is based on the combined usage of intellectual data analysis methods and advanced interactive visualization techniques. It should be noted that data analysis and data exploration are not the same. The user knows in advance what he/she is looking for, and in data exploration, the user does not. Data analysis implies deep understanding of the structure of the data, while data exploration is aimed at uncovering the general structure of the data. This paper is mostly focused on data exploration issues related to big data volumes.

HEP computing metadata as a test ground for InVEx

InVEx basics

The Level-of-Detail generator with nested grouping and clusterization

An instance of InVEx data exploration

Conclusion