PDC 2019: Award
Yesterday, I received the “Process Discovery Contest Award” for my winning submission to the Process Discovery Contest 2019. Many thanks to Massimiliano, Benoit, and Josep for organizing the contest. Next year, I will help them :-).
Yesterday, I received the “Process Discovery Contest Award” for my winning submission to the Process Discovery Contest 2019. Many thanks to Massimiliano, Benoit, and Josep for organizing the contest. Next year, I will help them :-).
Version 6.9.91 of the LogSkeleton package includes a new log visualizer plug-in that allows the user to use any classifier for the log skeletons. So far, the log skeletons were restricted to the “concept:name” attribute, but this restriction has now been lifted. The screenshot below shows the result of using the “org:resource” classifier on the Read More …
In the picture above: Note that I have dropped the dots from the edges, as they would be on all edges containing other symbols, hence they would have no meaning. The following picture then speaks for itself: Or, with hyper arcs:
Today, I have released version 6.9.62 of the LogSkeleton package. In this release, the thresholds have been replaced by a single noise level which can be set to 0, 1, 2, …, 19, or 20 percent. Furthermore, the equivalence relation (the ‘colors’ of the activities) now also takes this noise into account. As an example, Read More …
Today, I have released version 6.9.56 of the LogSkeleton package. The new features in this release are: The example shows that:
Today, I have released version 6.9.51 of the LogSkeleton package. The new features in this release are: In the example shown, activity a precedes activity h in 88% of all cases (that is, of all occurrences of activity h). Furthermore, note that the Not co-existence constraint for activity s is not shown. This constraints can Read More …
Today, I have released version 6.9.44 of the LogSkeleton package. The new features in this release are:
Using a fully-automated variant of the Log Skeleton Filter and Browser, I have managed to classify 195 out of 200 traces correctly for the Process Discovery Contest 2016. Not that this is better than my winning “DrFurby classifier” approach, which classified 193 out of 200 traces correctly. Below, you will find the main log skeletons Read More …
The Log Skeleton has been extended with how often an activity may occur in a trace. These numbers are shown in the bottom right corner for every activity. Typical values are “1” (occurs exactly once), “0..1” (may or may not occur once), and “0..x” (may happen up to x times). See below for an example. Read More …
The Log Skeleton readability has been improved by merging some of the arcs and by possibly introducing hyper arcs (to avoid clutter). The figures below show the effect on the log 10 of the Process Discovery Challenge 2017. Log Skeleton Browser As the computation of the hyper arcs may take considerable time, by default hyper Read More …