GreyCat: Efficient what-if analytics for data in motion at scale

Over the last few years, data analytics shifted from adescriptive era, confined to the explanation of past events, to the emergence of predictive techniques. Nonetheless, existing predictive techniques still fail to effectively explore alternative futures, which continuously diverge from current sit...

Full description

Saved in:
Bibliographic Details
Published in:Information systems (Oxford) 2019-07, Vol.83, p.101-117
Main Authors: Hartmann, Thomas, Fouquet, Francois, Moawad, Assaad, Rouvoy, Romain, Le Traon, Yves
Format: Article
Language:English
Subjects:
Analytics
Computer Science
Data analysis
Data models
Data Structures and Algorithms
Decision analysis
Dimensional changes
Graph processing
Graphs
Information systems
Mathematical analysis
Multidimensional data
Operating Systems
Predictive analytics
Software Engineering
Time-evolving graphs
Unstructured data
What-if analysis
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Over the last few years, data analytics shifted from adescriptive era, confined to the explanation of past events, to the emergence of predictive techniques. Nonetheless, existing predictive techniques still fail to effectively explore alternative futures, which continuously diverge from current situations when exploring the effects of what-if decisions. Enabling prescriptive analytics therefore calls for the design of scalable systems that can cope with the complexity and the diversity of underlying data models. In this article, we address this challenge by combining graphs and time series within a scalable storage system that can organize a massive amount of unstructured and continuously changing data into multi-dimensional data models, called Many-Worlds Graphs. We demonstrate that our open source implementation, GreyCat, can efficiently fork and update thousands of parallel worlds composed of millions of timestamped nodes, such as what-if exploration. •Existing analytics fail to effectively explore the effects of what-if decisions.•We combine graphs and time series into multi-dimensional models (Many-Worlds Graphs).•These are able to update thousands of parallel worlds composed of millions of nodes.•We benchmark our Many-Worlds Graph open source implementation, called GreyCat.
ISSN:0306-4379
1873-6076
DOI:10.1016/j.is.2019.03.004