Mondrian is a general purpose statistical data-visualization system. It features outstanding visualization techniques for data of almost any kind, and has its particular strength compared to other tools when working with Categorical Data, Geographical Data and LARGE Data.
All plots in Mondrian are fully linked, and offer various interactions and queries. Any case selected in a plot in Mondrian is highlighted in all other plots.
Currently implemented plots comprise Mosaic Plot, Scatterplots and SPLOM, Maps, Barcharts, Histograms, Missing Value Plot, Parallel Coordinates/Boxplots and Boxplots y by x.
Mondrian works with data in standard
tab-delimited or comma-separated
ASCII files and can load data from R
workspaces.
There is basic support for working directly on data in Databases
(please contact me for further info).
Mondrian is written in JAVA and is distributed as native application
(wrapper) for MacOS X and Windows. Linux users need to start the
jar-file. The latest version can be downloaded here.
For questions please refer to mondrian@theusRus.de. Bugs may be submitted to the bug-tracker as well as per e-mail.
News:
- (01/25/10)
Version 1.1 is out.
The major change is the ability to load dataframes directly from R workspace files into Mondrian,
simply by choosing the desired dataframe.
The color scheme has been improved (i) the color brushing colors are taken from the color brewer and (ii) there is a general scheme called "Daltonian" for those in need.
Furthermore this version works seamlessly with Java 6 on all platforms, improves queries in PCPs once again and fixes several bugs and adds minor feature requests.
(Filed bugs fixed since last release: 4, 27, 41, 44, 69, 75, 77, 78, 103, 112, 113, 114)
- (12/22/09)
The newsletter
article slipped in, such that version 1.1 is now sheduled for January -
Merry Christmas and a Happy New Year!!
- (09/10/09)
The slides
which go with the book "Interactive
Graphics
for
Data
Analysis – Principles and Examples" can now be
found on the web.
They should be a fruitful resource for everybody who wants to learn how to use Mondrian.
|
 |
- (05/24/07)
we have a
Bugzilla based bug-tracker
online. Please submit all bug reports and requests there - feel free to
e-mail after submission as well ...
Mosaic plots in Mondrian are fully interactive. Interactivity includes the standard operation as described in the convention section (except zooming), plus the reordering of the variables in the plot by using the 4 arrow keys. Use <meta>-r to rotate the direction along which the last variable in the plot is split.
Use <meta>-+ and <meta>-- to add and delete interactions during the modeling process using the ModelNavigator. During the model process one may want to use the plot option to show the expected values of the model and not the observed.
The picture below shows an example of the Titanic dataset, which includes information on the class (1,2,3 crew) age (child, adult) and gender (male, female) of the passengers. Surviving passengers are highlight.
Although there are no labels to decode the cells, the order of the variables is given in the title of the window. Using the interactive interrogation it is very easy to query the cells. In this static representation the knowledge of the fact that there are no children and hardly any women in the crew, should be sufficient to decode the plot.
Example of a rotated Mosaic plot, i.e. first variable is split along y not x!
Additionally Mondrian features 4 variations
of mosaic plots. The Figures below show the same data from the cars
data set, in all five possible variations. Use the pop-up menu for the
plot options:
Observed values
Expected values
(according to current model)
Same bin size
Fluctuation diagram
These plots show the five
different variations of Mosaic plots. Whereas the first two options are
"real" Mosaic plots, the other three versions (same bin size,
fluctuation diagram and multiple barchart) are more useful to handle
only a few variables with many categories, which is the worst case to
handle for a standard Mosaic plot.
For the latter three cases, Mondrian plots the category labels for the
first two variables, since the categories are equally spaced and thus
can be placed decently.
By typing shift-up-arrow and shift-down-arrow, the size of boxes can be
zoomed. As soon as a box reaches its maximum size, it is red-framed to
indicate the incorrect size. Note that in the same bin size display all
boxes have a red frame, because their sizes do not correspond to any
group count. 
Multiple barcharts
Barcharts in Mondrian follow a vertical layout, not a horizontal layout. Thus the level-names can be printed in full length.
Besides standard selection and interrogation techniques, interactivity in a barchart comprises:
- Reordering of levels via drag &
drop
(use <alt>-click on a bar or its text to drag)
(Dropping between categories inserts, dropping on categories exchanges) - Switch between proportional width and height
- Sort levels by
- frequency
- name
- absolute highlighting
- relative highlighting
- Reverse current order
A barchart for the dataset on the Titanic passengers. First class passenger are highlighted.
(The same behavior can be found in the variables window - plus additional selection of the hits)
Value Plot
If the dataset has missing values, a missing value plot can be used to analyze the structure of the missing values (monotone missingness etc.).
The options of the missing value plot are similar to those of a barchart (sorting etc.)
Missing values MUST BE CODED AS "NA"!!
Whenever a dataset provides information on polygons, Mondrian can draw interactive maps of this geographical reference.
A corresponding data record must be provided for each polygon defined in the dataset. Different polygons might point to the same data record, but multiple records to a set of polygons are ignored.
Maps offer the standard selection and interrogation techniques. Additionally the standard zooming function of Mondrian is enabled.
All maps have a popup-menu at the top to
create
a choropleth map of any of the variables; including
alphanumerical
variables.
Further options include:
- Color schemes
- "white to black"
- "red"
- "green"
- "blue"
- "blue to red"
- "blue to white to red"
- if R and Rserve are installed
- "heat"
- "terrain"
- "topo"
- Invert color scheme
- Assign color linearly, normalized, or by rank
- Limit minimum or maximum to a specific
value
 |
 |
| grayscale | green color scheme |
 |
 |
| blue to red color scheme | normalized blue to red
color
scheme |
 |
 |
| normalized blue to white to red scheme | linear blue to white to
red with limit to 30 |
Six choropleth maps of the five Midwest states, shaded according to educational status
The saturation (more precise the alpha) of boundaries can now be changed with the right arrow key (more saturation) and left arror key (less saturation).
This can change the perception of the map drastically, so make sure to test it out!
US County Map with full saturation:
US County Map with reduced boundaries:
Note the extreme difference of the maps! (We call this technique map-martinizing…)
Coordinates
Mondrian implements parallel coordinate plots for any arbitrary number of variables. Alphanumerical categorical variables are displayed equally spaced according to the currently defined order. Numerical variables are scaled according to their actual numbers.
Besides standard selection and interrogation techniques, interactivity in a parallel coordinates comprises:
- Reordering of the variables via drag
&
drop (use <alt>-drag)
- Switch between uniform and individual scale (also for selected subsets of axes)
- Adjust α-level of lines via context-menu or the left and right arrow key.
- Select Axes by a single click on the axis name and use:
- BACKSPACE to delete this axis from the plot
- <meta>-I to invert the axis
- Type PAGE-UP or PAGE-DOWN repeatedly to
cycle through the minimum number of ordering to see ALL adjacencies of
the variables in the plot.
(Note:
For k variables we need [(k+1)/2] permutations as shown in Ed Wegman's 1991 paper)
- HotSelection only shows the currently selected points in PCPs and PBPs
- Crop Selection removes the currently selected lines from the PCP. Subsequent crop operations allow to peel a data set. (does only work for lines)
A parallel coordinate plot for the olive oil data.
A parallel Box plot for the olive oil data.
The align menu offers means to align/center the axes
according to:
The last four settings use +/- 3 sigma/IQR as scaling. Use arrow key <up> and <down> to increase or decrease the scale. |
 |
(Parallel) Boxplots y by x only include a single variable, split by a second categorical variable. To invoke a boxplot y by x select the continuous variable to plot and the categorical variable to split by and select 'boxplots y by x' in the 'Plots' menu.
Manual reordering of the classes is only
possible
by reordering the levels in a corresponding barchart.
The context menu offers to automatically
sort the levels by either median or IQ-range (and to reverse a current
ordering.)
Parallel boxplots for the cars data set - heavy cars selected.
Scatterplots offer the basic interactions. Data can be selected and highlighted.
In contrast to most other plots in Mondrian, scatterplots offer axis, showing the maximum and minimum as orientation.
Interrogation methods inside scatterplots
operate on three levels. The first level is a simple overview of the
position of the cursor, which is displayed as projection onto the x-
and y-axis. This interrogation is invoked by simply pressing the
altenate key. A pressing control invokes the second level of
interrogation.
A tooltip is presented with the data of the variables in the plot
closest to the curser. If more than one point is found at the same
distance, a list of the cases is presented in the pop-up. Holding shift
and control shows the information of the
selected variables in the variables list.
 |
 |
 |
| orientation | objects | extended |
Three
levels
of
queries
in a scatterplot
 |
Example of an image query which shows a
photo of the car, which turned out to be an outlier in the MDS. In many cases it is very helpful to get additional information of an object, which might only be captured in an image, e.g. a chemical structure or a movie poster. Mondrian allow to specify the URL to the image location for each case. This can be an entire URL for each case, or a URL composed out of a common part and a case specifc part. The common part of the URL is coded in the column name, the case specific part is the entry in the column entry. |
| The format is as follows: A column holding an image URLs must start with '/U'. If there is a common part, it follows after the prefix. The position where the individual entry of the case goes is enclosed with '<' and '>'; this also defines the column name. Example: |
|
When Rserve and R are installed, scatterplots can be enhanced with scatterplot smoothers. Currently the list of smoothers comprises:
- least square regression (does not need R)
- loess smoother
- regression splines (with confidence intervals)
There is an option to either compare the highlighted smoother to all data or to the complement.
SPLOMS (ScatterPLOtMatrix) in Mondrian are "only" a collection of standard scatterplots, efficiently arranged in a single frame. Thus it has the disadvantage that all keyboard shortcuts apply to all panels simultaneously, but the advantage that each panel is a full featured scatterplot.
Hint:
SPLOMs are quite effective for a quick 2-d overview, but are very inefficient when working with more then just a few variables. In this case, parallel coordinate plots are far more effective.
The most crucial point in plotting histograms is to choose the ''right'' origin of the first bin and the ''right'' number of bins. Since there exists a vast amount of rules and hints, what ''right'' means under different assumptions, the most important interactive manipulation inside histograms is changing origin and number resp. width of the bins. This is done by pushing the arrow keys (up and down changes the number of bins, and left and right moves the origin). In order to keep the visual distortion as small as possible, the scale of the histogram axis is not updated during the interactive reparametrization. Obviously the y-scale must then represent probabilities and not counts.
<meta>-0 fits the y-scale after the bin-width has been changed.
The context menu allows to set fixed bin width and origins, either by using the suggested values or by entering arbitrary values.
When Rserve and R are installed, density estimation will enhance the histograms greatly.
(The right plot shows actually a CD-plot)
Histograms can now be weighted. Select two continuous variables (the weights usually should be positive, although Mondrian will not complain about negative weights) and choose weighted Histogram from the plot menu.
Above example shows a typical situation for weighting in a histogram. The left plot shows the distribution of %blacks for the US Midwest counties. The right plot is weighted with the total population, thus showing us the number of people living in areas with a certain % of blacks.
Selections in Mondrian can be made in two ways.
- Simple Selection
- Selection Sequences
Simple selections are performed as any selection in the operating system's desktop. A new selection replaces the current selection.
Holding down the <shift> key will combine the new selection with the currently selected data in XOR-mode.
Holding down <shift> and <alt> will perform a selection in extended mode, which is AND by default, but can be changed to OR in the Options menu.
When using Selection Sequences, any selection is recorded. The selection is represented by a transparent rectangle with 8 handles. Use any of these handles to resize the rectangle (slice) or click-drag the rectangle to move (brush). The popup-context menu on a selection rectangle will indicate the selection step and offer the choice of changing to a different selection mode (union, intersection, negation, xor), of deleting this step, or the complete sequence. Deleting a single step can also be performed by <backspace>. Use <meta-backspace> to delete the complete sequence. To query objects covered by a selection rectangle hold down the <shift> key to click trough the rectangle.
Selection Sequences can span across plots and more than just one selection can be made per plot. To keep track of the selections made, all selections are annotated in the windows menu, just behind the window title, i.e. "Scatterplot(x,y) [2] [4]" tells us that selection steps 2 and 4 have been made in the scatterplot of the variables x and y.
Use <meta>-a to select all cases.
Note: Deleting all selections is not limited to the current plot window.
Brushing
Wheras selections are a more transient technique to mark a subgroup of interest, color brushing persistently assigns colors to cases. There are three ways to define persistant colors in Mondrian
- Definition in a barchart or a mosaic
plot
via <meta>-b sets a descrete scheme
- Definition in a histogram via <meta>-b sets a
continuous "rainbow"-scheme
- Individual colors may be assigned using <meta>-1 to <meta>-9
 |
 |
| LEFT: In the map of the German voting destricts, the party who won the most votes in 2009 was colored. TOP: The barchart shows the same coloring for the different states. |
To clear all colors use either the (context) menu option or press <alt>-<meta>-b.
Although looking very useful at the first sight, one should keep two very critical issues with color-brushing in mind:
- Permanently defined colors alway interfere with the highlighing color and may cause confusion
- Wheras overplotting is no problem for
highlighting (it is alway plotted on top) the overplotting issue with
multiple colors can not be resolved satisfactory (α-transparency won't work here too well)
The key to a smooth and efficient user interface are conventions. Once the user learned the basic set of operations like, selections, interrogations, zooming and alteration, she/he can perform these operation within any plot.
In an interactive graphical system, possible interactions can be performed by mouse and keyboard. Since JAVA programs are not bound to a specific platform, Mondrian tries to only makes use of features, which can be found on all platforms. There are some restrictions like one-button-mouse for most MAC-users (Steve give us the right button!!). The most commonly found modifier keys are SHIFT, CONTROL, ALTERNATE and META. CONTROL is blocked as the popup-trigger on the Macintosh, META abused under Windows and ALT blocked by many window-managers under Linux.
The interactions in Mondrian are assigned as follows:
- click and drag -> create a selection (rectangle)
- META-click and drag -> zoom in/out
(is
middle-click (wheel) and drag on Windows & Linux)
- META and mouse-over -> query object
(use META-SHIFT to get extended query)
- popup-trigger on background -> alter the plot setting
- ALT-click and drag -> reorder objects
The α-channel can be used to specify the transparency of an object painted. This is very useful, when plotting really many objects , which would result in heavy overplotting. Thus the density of objects can be easily displayed.
The figures below show an application using the well known "pollen" dataset.
There is a new option to invert the density scheme. (Does not work on Windows)
Although Mondrian was not designed to support statistical modeling of datasets, a graphical modeling technique for categorical data using Mosaic Plots is built in.
The so called ModelNavigator allows a stepwise graphical modeling of loglinear models.
The ModelNavigator basically inverts the usage of graphs and models. Whereas packages like R or S-Plus usually assume a model, for which diagnostic plots can be plotted, the approach in Mondrian starts with a graph, to set up a model, and uses the statistical measures as diagnostics, to reinforce the graphical implications.
For a more precise description of this
technique
see the paper on Visualization
of Loglinear Models.
Mondrian supports the standard ASCII data format, which consist of a header of variable names, and tab-delimited columns.
Numerical and alphanumerical data may be used. See example below:
U.S. Buick Estate Wagon 16.9 4.36 155
U.S. Ford Country Squire Wgn 15.5 4.054 142
U.S. Chevy Malibu Wagon 19.2 3.605 125
U.S. Chrysler LeBaron Wagon 18.5 3.94 150
U.S. Chevette 30 2.155 68
Japan Toyota Corona 27.5 2.56 95
Japan Datsun 510 27.2 2.3 97
U.S. Dodge Omni 30.9 2.23 75
Germany Audi 5000 20.3 2.83 103
Sweden Volvo 240 GL 17 3.14 125
Sweden Saab 99 GLE 21.6 2.795 115
France Peugeot 694 SL 16.2 3.41 133
...
The mode of non-numerical variables can be set interactively:
 |
Double
click (or use <meta>-T if you need to change more than one
variable) a continuous
variable ( ) to change it to be categorical ( ). Vice
versa double click to change to .Alpha numerical variables are tagged as  and can't be
changed to any other type. To get a fast and effective overview of which variables have missings, there are 'white'-versions of the icons of all three types, i.e.  ,  and 
indicating at least one missing in the particular variable. |
When working with very many variables, just type the prefix of a variable name to search for a particular variable and to make the variable window select the hits and also scroll to the first hit.
Polygon Data must be stored in a separate map file
The format for map-data
The dataset must include
one variable of references,
the polygons can refer to. This variable must start with /P. If a dataset refers to a polygon, there must be an
empty line after the data matrix followed by the relative path+filename
to the file containing the map data.
In the map file, each
polygon must be defined as follows:
It must start with a header like
id \t name \t n
where id is the matching id from the reference
variable. Name can be any arbitrary string naming the polygon. n is the
number of points in the polygon.
This header is followed by x and y coordinates
defining the polygon - separated by tabs, one pair per line. The first
and last coordinates must match, i.e. the polygon must be closed.
An example for Union
county:
...
-1.3050 0.7141
1761 /Pnew jersey,union 25
-1.2981 0.7112
-1.2997 0.7100
-1.2995 0.7097
-1.2990 0.7099
-1.2988 0.7098
-1.2991 0.7094
-1.2992 0.7090
-1.2999 0.7086
-1.2985 0.7088
-1.2969 0.7089
-1.2969 0.7088
-1.2964 0.7087
-1.2954 0.7088
-1.2951 0.7090
-1.2947 0.7095
-1.2945 0.7095
-1.2942 0.7100
-1.2942 0.7102
-1.2945 0.7103
-1.2949 0.7103
-1.2956 0.7106
-1.2965 0.7108
-1.2970 0.7107
-1.2976 0.7108
-1.2981 0.7112
1762 /Pnew jersey,warren 33
-1.3112 0.7149
...
Mondrian supports loading data directly from R workspaces. To do so, one only needs to specify the R workspace file (in most cases it might default to .RData). Once the workspace file is chosen, Mondrian lists all dataframes within this workspace. Selecting the desired dataframe will dump a temporary file from R and read this file into Mondrian.
Connections
The development version of Mondrian allows the connection to databases via the JDBC interface.
Currently this type of connection, which leaves the data entirely inside the database, is under further development and thus not released with the latest releases.
The figure below shows the database connection dialog:
Here are some sample data sets, which are ready to load and test with Mondrian (make sure to save the link directly to preserve the tabs!):
Titanic
Data set on the 2201 passengers of the Titanic. Pure categorical with data on class, gender, age and survival.
Pollen
Fake data set with hidden feature, which is easily found with α-channel
features.
Olive
Oils
Data set on Italian olive oils. Several fatty acids have been measured
which allow to separate the different regions from Italy.
Berlin
(old map format)
Election and socio-economic data on the city of Berlin, shortly after
the Berlin Wall was torn down. Includes a polygon of the districts of
Berlin.
US
Election
2004 (new map format)
Data on the 2004 US presidential election. Includes polygons of the Counties of the US.
Data courtesy of GeoVISTA (http://www.personal.psu.edu/users/a/c/acr181/election.html)
Mondrian features a preferences dialog, to set your favorite background and highlight color. Five schemes are preset. If you have some other intriguing color scheme, please let me know to integrate it.
You may save your custom scheme if none of the defaults suits you.
By downloading any version of Mondrian, you accept the following license:
Copyright (c) 1997-1998 AT&T Labs Research,
2002-2006 University of Augsburg.
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 3 of the License, or (at
your option) any later version.
This program is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, see <http://www.gnu.org/licenses/>.
Read-only svn-access to the source code:
svn://svn.rforge.net/org/trunk/rosuda/Mondrian/
Binaries for Windows, MacOSX and Linux:
1.1 as of 1/25/2010.
Windows (exe-file)
UNIX (JAR-file)
Mac OS X (Disk-Image containing application and demo data)
Changes:
- Load data directly from R workspace files
- New color schemes
- Compatible with Java 6 on all platforms
- Very many bug fixes and minor features added
1.0 as of 12/18/2008.
Windows (exe-file)
UNIX (JAR-file)
Mac OS X (Disk-Image containing application and demo data)
Changes:
- Autostart of Rserve under Windows and Linux
- Searchable variable list window
- Missing value plot is compatible to color brushing now
Changes in Version 1.0 beta12 as of 08/29/2008.
- Support for Rserve 0.5+
- Fixes and clean-ups
Changes in Version 1.0 beta11 as of 03/19/2008.
- Image can be used in extended queries for URL variables
- New color scheme in maps
- Search in barcharts by typing a prefix of the level
- Fixes and clean-ups
Changes in Version 1.0 beta10 as of 12/16/2007.
- More consistent menu entries and menu labels for plot windows
- A 'Open Recent ...' menu entry
- Indication of missingness in the variable window icons
- Window sizes can now be set in the scale dialog box
- Censored zooming in barcharts (shift up/down-arrow) consistent with mosaic plots
Changes in Version 1.0 beta7 as of 05/13/2007.
- Rserve start-up compatible with Rserve for R2.5.x
- SPLOMs are available now (for those who like'm ...)
- histograms are more consistent now (weighted histograms support densities (needs Rserve), spinograms now work at any zoom level)
- better scaling and queries in parallel boxplots (still incomplete)
- several fixes and enhancements ...
Changes in Version 1.0 beta3 as of 10/31/2006.
- simple transformations (+, *, -, /, log, 1/x, ...)
- selection order of variables in variable window is reflected in all multivariate plots!
- many minor fixes and enhancements ...
Changes in Version 1.0 beta1 as of 05/24/2006.
- new much faster loader (note: maps are now expected to be in a separate file)
- missing values (coded as "NA") are suported in all graphics
- missing value plot can be used to investigate the structure of the missing values.
- custom scaling (<meta>-j), scatterplot only, other plots to follow
- color brushing (<meta>-b) in barcharts, mosaic plots and histograms (rainbow)
- <meta>-1...9 sets persistent colors for the current selection
- derived variables from selection- and color-state
- painting, via "OR"-mode in the first selection step of a selection sequence
- many minor fixes and enhancements ...
Changes in Version RC 1.0m as of 11/29/2005.
- Using 1.4.x JVM on all platforms.
- '<-' and '->' can be used to change the saturation of boundaries in maps.
- "Boxplot y by x" is now a separate menu item.
- Levels can now be sorted in boxplots y by x according to median or IQ-range.
- Plotting of 2-dim MDS (input is not carefully checked yet)
Changes in Version RC 1.0f as of 04/06/2005.
- Queries are now implemented via ToolTips.
- Further improvements to Parallel Coordinate Plots. See section for details!
- Maps now feature six different color schemes for shading choropleth maps.
- Under MacOSX you can now drop files on Mondrian to start the application and load the data.
- If you have R and Simon's Rserve installed on your machine, you find new features in
+ Histograms
+ Scatterplots
Changes in Version RC 1.0 as pf 09/24/2004:
- Vast improvements to Parallel Coordinate Plots. Se section for details!
- Printing works via <meta>-P in all plots. In MacOS X use "Preview" to save as PDF.
- Additional sorting options in Barcharts.
- Histogram parameters can now be set manually as well.
- Choropleth maps can now be inverted and colored by rank.
- Yet another update to the L&F of selection sequences.
Changes in Version 0.99a as of 03/11/2004:
- an updated version of selection sequences. See the section for details.
- "window" menu and more intelligent window placement
- new controls to set width and origins in histograms
- zooming for all platforms (use middle mouse button on all other machines than mac)
Changes in Version 0.99 as of
11/18/2003
- Three new variations of Mosaic plots (same
bin size, fluctuation diagram and multiple barcharts)
- Automatic sorting of axes in a parallel
coordinate plot
- Use meta-R to switch the splitting direction
of the last variable in a Mosaic plots
- Inverse color scheme for density
highlighting in scatterplots
- Preference box to set highlight color and
background color
- Zooming under Windows is still delayed
because of a yet to be finalized major update on the interface
Changes in Version
0.98 as of 03/22/2003:
- Boxplots y by x. Just select a continuous
variable and a categorical variable and choose 'parallel boxplots' in
the plot menu.
- Regression lines in scatterplots (can be
queried)
- Highlight color is now red!
- Add lines in scatterplots by third variables
to visualize paths and other relationships.
Changes
in
Version
0.97a
as of 11/21/2002:
- oneClick selection is introduced, i.e. a
selection rectangle of size 0 will only select the clicked object, but
NOT create a corresponding selection rectangle (selection is only
temporary as with the select all feature (META-a))
- Bug fix in Scatterplots
- Update on selection rectangle appearance
Changes in Version 0.97 as of 7/12/2002:
- META-a will select all points in any plot now
- alpha-channel works in scatterplots (use arrow keys to change) and
parallel coordinates (via pop-up).
- scatterplot are automatically binned, if the dataset is really large
(can be overridden)
- interrogation in maps added
First public release 0.96 as of
4/9/2002
Note: The JAR-file can be started by a simple double click. Within
Windows, Sun's JRE or JDK must be installed (make sure that .jar files
are not associated with any decompression application), Mac OS X users
just smile.
The best reference to cite Mondrian - apart from the website - is 'the book'.
@book{1502124,
author = {Theus,, Martin and Urbanek,, Simon},
title = {Interactive Graphics for Data Analysis: Principles and Examples (Computer Science and Data Analysis)},
year = {2008},
isbn = {1584885947, 9781584885948},
publisher = {Chapman \& Hall/CRC},
}
There is also the JSS-Paper:
@article{Theus:2002:JSSOBK:v07i11,
author = "Martin Theus",
title = "Interactive Data Visualization using Mondrian",
journal = "Journal of Statistical Software",
volume = "7",
number = "11",
pages = "1--9",
day = "22",
month = "11",
year = "2002",
CODEN = "JSSOBK",
ISSN = "1548-7660",
bibdate = "2002-11-22",
URL = "http://www.jstatsoft.org/v07/i11",
accepted = "2002-11-22",
acknowledgement = "",
keywords = "",
submitted = "2002-07-11",
}
Starting Rserve
Getting the tiny warning message after starting Mondrian does only indicate that there is no connection to R. This will NOT harm Mondrian in its core functionality - one can happily live without the R connection.
Rserve is now a regular R package and can be installed as such. Right now, Mondrian will start Rserve automatically only under MacOSX. For the time being, Windows and Linux users need to start Rserve manually. Detailled instructions on how to do this can be found here.
File Formats
If you have trouble getting data loaded into Mondrian, load the file into MS Excel first, and check whether
- all variables have a name in the
header, i.e. the first row/line
(When saving data from R without omitting the row names, the "row.name" column has no heading name!)
- there are empty cells in the file.
Mondrian currently does not tolerate empty entries.
If your data is more complex or too large, you may do the similar "data washing" procedure using R. It is most easy when you use the JGR GUI, which has an "Open Dataset" menu item and allows to save the file from the object browser.
Starting a JAR File under Windows (if one can't use "Mondrian.exe" for some reason)
After an installation of SUNs latest JRE (or JDK), .jar-files can be started by a simple double click. If this does not work, the following two problems might be the cause
- There is no application associated with the suffix .jar. To change this, follow these instructions.
- Another application "grabbed" the responsibility for .jar-files - usually "WinRAR" - after the installation of the JRE. To change this, you must:
- Launch the WinRAR application
(Start -> Programs -> WinRAR -> WinRAR) - Once the application has been started, select "Options", "Settings".
- Select the "Integration" tab.
- This tab lists all associated file types for the WinRAR application. De-select the "JAR" type and click "Ok".
- Close the WinRAR application.
Other Problems
Please mail to either mondrian@theusRus.de or the mailing list stats-rosuda-devel, or submit your issue at the Bugzilla based bug-tracker.
Issues of general interest will be posted on this page.