A set of general-purpose Knowledge Discovery and Visual Data Mining application components for analysis and visualisation of multidimensional data in Excel spreadsheets. MDViewer, 3D Scatter, 2D Scatter and ParallelCoordinate application components (Stand alone, Networked and Smartdoc versions):
Knowledge discovery and visual data mining, which consist in extracting valuable and relevant knowledge from large volumes of data, have received much attention these last years. The approaches used for knowledge discovery are non-trivial and often domain specific, depending on the canonical mining primitives. The data patterns discovered are typically used in decision-making whether in business, in scientific research or other. While significant research has been done on knowledge discovery from large corporations, most of the approaches are related to numerical transactional data such as market-basket analysis, web activities, etc., thus very little has been achieved on visual data mining probably due to the complexity of integrating interactive exploration, visual user interface and data mining.
Visual Discovery can be even more valuable when it is combined with visualization technology that provides real time reporting with three-dimensional graphics. The SmartDoc Knowledge Discovery components transform spreadsheet data into visual metaphors-3D, colour-coded, interactive displays. These visual metaphors transform complex statistical relationships into clear visual information SmartDocs for the end user. 3D Scatter and Parallel graphs, for example, can be expanded, zoomed, panned and essentially animated embedded into dynamic electronic documents.
Four different SmartDocs in the area of Knowledge Discovery have been developed:
- Multidimensional (MD) Viewer
- 3DScatter
- Parallel Coordinates
- HTE
The MD-Viewer SmartDoc consists of three knowledge discovery components; 2D scatter chart component, a 3D scatter chart component and a ParallelCoordinates component. These three components visualize the same dataset and complement each other, further enhancing the understanding of the dataset.