DISQOVER 4.00 and the power of Visual Analytics
With the release of version 4.00 of our DISQOVER platform, we bring the power of Visual Analytics to your data discoveries. The deep integration with DISQOVER’s core linked data philosophy creates a unique offering, and allows you to endeavor data exploration journeys that were previously unthinkable.
What is Visual Analytics?
A seminal paper by Thomas & Coock(1) describes it as “the science of analytical reasoning facilitated by interactive visual interfaces” . Its aim is to assist the discovery of insights from massively complex data sets, in situations where even the questions are often not clear, by strengthening the synergy between two fantastic tools: computers and the human brain.
Visual analytics facilitates knowledge creation through rapid interaction cycles between computer-generated data visualizations and a human expert using domain knowledge, intelligence and creativity to interpret the visualizations, and actively steer the cycles.
Conventional model-based analytics sequence
● Sequential approach.
● Hypothesis required at an early stage.
● Visualization only at a later stage.
Visual analytics discovery cycle
● No fixed and clear a-priori hypothesis necessary.
● Visualization at an early stage, driving hypothesis creation.
● Multiple, rapid and interactive iterations during a single session.
How does it work in DISQOVER?
Visual analytics dashboards
In DISQOVER 4.00, you can create your own rich visual analytics dashboards using a wide range of chart types, ranging from simple bar graphs to sophisticated chord diagrams. Many of them, such as heat mapped pivot tables and bubble charts, are specifically designed to show interactions between several facets of the same data. Each chart is fully configurable and offers advanced features for real-time navigation and data selection.
All charts in a dashboard are dynamically connected via a visual analytics concept known as brushing and linking: if you select data on one chart (e.g. dragging a rectangular area on a scatter plot), the selected data becomes automatically highlighted on all other charts. By combining visual selection actions on different charts, you can drill down the data set to reveal and highlight interesting patterns.
Once you’ve created an interesting visual analytics dashboard, you can save it as a template and instantly re-apply it on any new data set that you encounter during future searches in DISQOVER. You can save as many custom visual analytics templates as you want, each of them focusing on a different aspect of your data. Templates can be stored for only you, or made available to other users within your organization.
Integration with linked data
We created a deep, bidirectional integration between the new visual analytics dashboards and DISQOVER’s powerful semantic philosophy and unique linked data features.
You start your journey by entering a search term and perhaps follow links to related data. At any moment you can take the found set of instances to a visual analytics dashboard, where you can start mining for trends using the advanced charts, and further drill down the data set with the interactive selection tools.
But what makes it really exciting is that, in a visual analytics dashboard, you can use the instances present in your drilled-down data set to follow links to related data . And nothing prevents you from exploring that linked data again in a new visual analytics dashboard.
What does this mean for you?
DISQOVER is known to be the tool of choice for discovering information in complex, hybrid and linked data sets originating from a broad range of sources. Version 4.00 now also allows you to discover quantitative trends in the same data, using the power of visual analytics tools. And everything is seamlessly integrated into a single user-friendly platform, relieving you of the worries of having to transfer data from one environment to another.
In this way, DISQOVER further enforces its unique proposition of being the ultimate guide for all knowledge discovery journeys through your data.
At ONTOFORCE we are thrilled about this new evolution of our product. We hope you are, too!
1 “Illuminating the Path. The Research & Development Agenda for Visual Analytics”, J. J Thomas & K.A Coock, 2005 (https://vis.pnnl.gov/pdf/RD_Agenda_VisualAnalytics.pdf )