To prevent spam users, you can only post on this forum after registration, which is by invitation. If you want to post on the forum, please send me a mail (h DOT m DOT w DOT verbeek AT tue DOT nl) and I'll send you an invitation in return for an account.

Interactive Data-aware Heuristic Miner - query/filter problems

carvalhombcarvalhomb Posts: 2
edited June 3 in - Usage
Hello, all,

when trying to use the Interactive Data-aware Heuristic Miner plugin interface, I'm running into a strange behavior.

I'm trying to run queries against a log that has many trace-level attributes. I only get zero results when querying for any attribute that is not literal typed. However, I'm sure that the queries should yield results, as they run fine when I try the very same queries using Felix Mannhardt's Log Enhancement's log visualizer. Also, when the trace-level attributes are moved to event-level, the queries also yield the expected results.

Has anyone noticed this? Is it a bug? Or should I be writing the queries differently when querying for trace-level attributes?

Thanks in advance!
Regards,
Maira
Post edited by carvalhomb on
Tagged:

Comments

  • fmannhardtfmannhardt Posts: 57
    HI Maira, the underlying query engine should be the same, but it is wired up differently. Could you give an example query and log that shows the issue?
  • carvalhombcarvalhomb Posts: 2
    Hi Felix, thanks for the reply!

    I cannot share my dataset, unfortunately, but I'll try to explain better what I mean and give some examples.

    I have an event log imported from a CSV. Case-level attributes in my dataset are columns that start with "CASE_". I converted the log from CSV to XES, and then used your plugin to move attributes from event to trace level. I confirmed in the event log visualization that the attributes are showing at the case level, not at event level.

    The query I am trying to perform that works well in this setup looks like this (CASE_CLUSTER is a literal typed attribute):
    CASE_CLUSTER="Cluster_01"
    I press enter and I see the text under the search field: "using 70/1391 traces, 467/15769 events"
    Then, if I try:
    CASE_LENGTH>1
    (CASE_LENGTH is a continuous typed attribute)
    I get "using 0/1391 traces, 0/15769 events", even though I'm 100% sure I have entries that have CASE_LENGTH larger than 1. In fact, the exact same query in the Log Visualizer plugin gives me 746 traces. 

    Another example:
    CASE_TYPE=1
    (CASE_TYPE is discrete typed)
    also gives zero results in IDHM, but 129 traces in the Log Visualizer plugin.

    All the above queries give me the expected results if I don't move attributes from event to trace level.

    Does that make sense?

    Best regards,
    Maira
Sign In or Register to comment.