dbslice

Interactive, hierarchical, database-driven plotting

Demos »

dbslice is a JavaScript library for web-based data exploration. A collection of data instances (Tasks) is interactively filtered, based on high-level descriptors, to obtain a selection of Tasks for deeper exploration. The selected Tasks are then compared using scatter, line, contour, 3D surface plots, and Augmented Reality rendering, generated on demand.

dbslice is built on the d3.js, crossfilter.js, three.js and ar.js libraries.

Interactive filters

Filter a large database of Tasks using high level descriptors of each case (Type, date ...) or by using value metrics (Cost, size, location...). Detailed data from the selected Tasks can then be extracted from the database and analysed.

Comparative plots

Data are organised by type (point, line or surface), property and Slice (location or other subset). Point, line and contour plots (d3.js) are interactively added, refreshed and removed from the view.

WebGL performance

3D surface plots (three.js) use WebGL acceleration for smooth rotation and scaling. Several surfaces can be displayed in a 3D view, as in this multi-stage compressor example. Models can also be viewed in Augmented Reality (ar.js).