ANALYST: Advanced Analytics on Smart Data

Improved processing power and decreased costs of sensors and data storage has created a data deluge: so much data is generated that extracting useful information from the data is like finding a needle in a haystack. In ANALYST, we will expand the state of the art of Big Data Analytics, turning Big Data into Smart Data using innovative techniques for Locally Refined Spline modelling.

The challenge of managing, investigating, and visualizing big datasets is not new in the fields of Science, Technology, Engineering and Mathematics (STEM). From seismic surveys of new oil fields, through Computational Fluid Dynamics (CFD) simulations of waterpower turbines, to large-scale meteorological forecasts, we have always faced limitations in computing power and storage capacity. Recent developments in BDA offer a new set of tools for overcoming these challenges. However, there are significant new challenges that arise from the structural differences between most STEM data and the unstructured textual data typical in classical Big Data applications. STEM data such as simulation results and sonar surveys has a spatial structure, displays complex behaviour, and comes in inconveniently large chunks.

In order to address these challenges, ANALYST will use Locally Refined (LR-) spline data modelling to turn Big STEM Data into Smart Data. In early implementations of LR-spline algorithms in 2D and 3D, we have seen their potential as compact interactive models for visual and quantitative analytics on big spatial datasets. The data structure of LR-splines is directly related to the spatial distribution and local variation in the data, enabling efficient hardware-accelerated interrogation and visualization. By splitting the data into tiles, approximating each tile independently, and stitching these approximations back together, we have an extremely versatile and parallelizable approach, well suited for Big Data infrastructures. This tiling-and-stitching procedure can be achieved for LR-spline models in any dimension. We can therefore include time and other relevant variables in a compact, interactive, multi-scale, higher order, locally refined model.

ANALYST will provide the research platform to bring the theoretical foundation of LR-splines up to a level where their full potential can be explored, combining BDA and AI to provide advanced analytics on the LR-spline model. While the focus will be on data from applications in the STEM fields, the resulting algorithms have a wider applicability, providing highly scalable complex modelling tools for Big Data Analytics.

Combining Big Data, Artificial Intelligence and LR-splines
ANALYST: Combining Big Data, Artificial Intelligence and LR-splines
LR-spline field visualizations
LR-spline model of a wind simulation at the proposed location of a telescope. Left and middle: transparent multi-scale visualization of the wind speed. Right: streamlines of the wind velocity. Data courtesy of the GiD and CRATOS groups at CIMNE.

Prosjektdeltakere

Heidi Elisabeth Iuell Dahl

Senior Research Scientist
Name
Heidi Elisabeth Iuell Dahl
Title
Senior Research Scientist
Phone
470 77 444
Department
Mathematics and Cybernetics
Office
Oslo
Company
SINTEF AS

Tor Dokken

Research Manager / Chief Scientist
Name
Tor Dokken
Title
Research Manager / Chief Scientist
Phone
930 58 710
Department
Mathematics and Cybernetics
Office
Oslo
Company
SINTEF AS

Georg Muntingh

Research Scientist
Name
Georg Muntingh
Title
Research Scientist
Phone
410 03 728
Department
Mathematics and Cybernetics
Office
Oslo
Company
SINTEF AS

Sverre Briseid

Research Scientist
Name
Sverre Briseid
Title
Research Scientist
Phone
402 23 390
Department
Mathematics and Cybernetics
Office
Oslo
Company
SINTEF AS

Vibeke Skytt

Senior Research Scientist
+47 924 63 527
Name
Vibeke Skytt
Title
Senior Research Scientist
Phone
+47 924 63 527
Department
Mathematics and Cybernetics
Office
Oslo
Company
SINTEF AS