The VisualMine Image Gallery

Welcome to the VisualMine Image Gallery, a collection of 3D visual analysis examples. All the following examples have been obtained on customer data, interactively generating the visual representations using VisualMine Professional. Examples are provided covering:

• Multiple variables correlation analysis

• Time series analysis

• Pattern identification

• Anomalies detection

• Visual cluster analysis

• Geographic analysis

Multiple variables correlations - 1

Viewer: Scatter 3D with cutting planes.

Use: Correlations among up to 7 contemporary variables are visualised.

Example: Different variables describe the behaviour of bank branches for one of the largest Italian banks. The behaviour is described with respect to financial, commercial, personnel and infrastructure cost data

Notes: This technique can show correlations among up to 20 contemporary variables.

Multiple variables correlations - 2

Viewer: Scatter 3D + interpolation + isolines and isosurfaces

Use: The distribution of one variable is analysed with respect to other 3 or 4

Example: Employed in anti money laundering. Distribution of bond trading activities with respect to the number and volumes of accounts in bank branches. The surface connects a cluster of branches trading the same amount of bonds in a given month.

Example: Interpolated fields are also probed using isolvolumes and orthoslices.

Multiple variables correlations - 3

Viewer: Bars 3D with joint frequencies.

Use: Anomalies detection, correlation analysis, basket analysis

Example: Joint frequencies for two variables, x and y, are displayed in the z dimension. The two variables correlations are displayed, as well as high/low frequency combinations.

Multiple Variables Correlations - 4

Viewer: Torus 3D

Use: Multiple variables on multiple entities are displayed.

Example: A detailed comparison between a limited number of entities is obtained. In this example each colour is associated to a bank branch, each disk representing one behavioural variable.

Time Series Analysis

Viewer: Bars 3D (stripes view), with animation

Use: 3D time series visualisation

Example: Time series for two variables are visualised (mapped on colour and height) for a set of bank branches. In addition to this type of graph, all the 3D graphs can be animated on any variable.

Anomalies Detection - 1

Viewer: Bars 3D (planes, surfaces) with thresholds

Use: Visual identification of anomalous entities (by position, height and/or color)

Example: Anomalous entities are easily recognised. The information pickup facility, on mouse click, allows to easily access the detailed information on the represented entities.

Notes: Thresholds can be applied on two different variables at the same time.

Anomalies Detection - 2

Viewer: Scatter3D

Use: Visual detection of anomalous entities (position, dimension, color)

Example: Employed in anti money laundering to detect behavioural anomalies on a large number (hundreds or thousands) of financial intermediaries. Employed also to detect anomalies within large organisations.

Analysis of Clusters

Viewer: Scatter 3D with Solid Contour and thresholds

Use: Understanding clusters distribution

Example: The distribution of clustered data is analysed with respect to variables included/not included in the clustering process. Each color represents a cluster: the graph shows the distribution of the clusters related with three variables (x, y and z). Used in market analysis and customer segmentation.

Pattern Identification

Viewer: Scatter 3D, with interpolation and Solid Contour

Use: Pattern Identification

Example: Data distributions in a Scatter 3D graph are studied using solids to represent clusters. Distribution patterns are then understood and analysed.

Geographic Analysis - 1

Viewer: 3D Maps

Use: Geographical analysis

Example: Traditional territorial analysis by means of thematic maps. Here data are studied on three levels of detail (municipality, province and region) at the same time. Employed in anti money laundering as well as in marketing.

Geographic Analysis - 2

Viewer: Map Flow Visualisation

Use: Visualisation of relationships among geographic entities

Example: Employed in anti money laundering to understand macroscopic money flows and to detect anomalous situations. Useful also for traffic analysis for Telecom companies. More information on relations can be picked up with a mouse click.

Send mail to webmaster@ais.it with questions or comments about this web site.

Copyright © 1997 Artificial Intelligence Software SpA
Last modified: February 12, 1997