InfoScope Distiller

InfoScope Distiller (part of InfoScope Professional) allows you to process your data and produce files that can be opened using InfoScope. It has a wizard-like interface that walks you through the loading of the data and then through the configuration of the various views that InfoScope provides.

To load your data, first launch InfoScope Distiller and then select a file containing the data to be processed.

Various popular formats to store tables are supported, such as tab-delimited (.txt) and comma-separated (.csv), as well as the native InfoScope file format (.mis) that InfoScope Distiller produces.

Each row in the table corresponds to an observation. The columns represent the variables that describe the observations. The first row of the table must contain a unique name for each variable.

As an example, you can have a look at the Prices and Earnings around the Globe 2006 dataset in tab-delimited format.

In this dataset, there is first a header line describing the variables in each column. Then each of the following rows describes an observation, a city in this case.

Datasets in tab-delimited format can be easily produced using e.g. Excel. When doing "Save as..." just choose the appropriate file format. Excel will warn you that some information (formatting, colors, formulas, etc.) will be lost in this format, but that's ok, we don't need it for InfoScope.

When your file has been loaded, you can verify that the data has been correctly read in.
Two tables are provided for this. One table shows the raw data, and another one the statistics.

Please check that the minimum and maximum values, as well as the number of missing values for each variable are correct.

Then set the target language for the interface and define the key element that your data contains. The latter will affect the labeling in the status bar at the bottom of the InfoScope user interface.

You can define the variables that should be represented as lists by inserting variables from the left into the table on the right. These are typically the categorical variables describing your observations. They will be displayed in the upper part of the InfoScope user interface.

By default, the lists are labeled with the column names, but they can be changed by simply editing the labels in the list on the right.

Numerical variables can be inserted to be part of the graphical parallel coordinates display, provided in the lower part of the InfoScope user interface.

Variables can be organized into groups that can be collapsed by the user in the interface to allow a large number of variables to be displayed concurrently.

Groups can be assigned a color. To change the color, first make the color column visible in the table on the right, by clicking on the little column chooser in the top right corner of the table. Then click on the little square color swatch for a group header, to open up a color chooser.

You can set the minimum and maximum values to any value by simply editing the numbers in the table. Right-clicking on a label brings up a context menu with additional options for setting the min and max values. For example, you can equalize the ranges of several variables by selecting all of them, and then choosing the "Equalize min/max..." option from the context menu.

The labels of the axes are the variables names by default, but can be changed by editing them in the table on the right.

Both categorical and numerical variables can compose the table view. Insert the variables available on the left into the list on the right and arrange the order of the columns that will compose the table. Different renderers can be defined for the table view cells.

A geographical map can made visible in the user interface. Background maps can be loaded in Shapefile format. Either polygons from the map can be linked to observations in the data, or points in decimal longitude/latitude degrees from the data can be used to interact with the observations.

Note: The following feature is in beta state. You can use it, but results may be mixed. In the next version the similarity map feature will be properly working and documented.

Similarity maps can be computed and organized into themes. To produce a similarity map, select the variables that should be part of the similarity metric.

Choose a file to store the data for InfoScope.