Hi,

you can find a number of datasets in the UCI Machine Learning repository: https://archive.ics.uci.edu/ml/datasets.html

best,

Frank

Thanks so much for your reply. However, I want a dataset specifically for process mining applications. In which the dataset is basically a log containing low-level events, where each event has a name, timestamp, originator, and the class label is the corresponding high-level activity.

Thanks

Hi Ghalia,

Perhaps if you have a sample dataset (that you can share) or a textual resource about your task, that would be much useful for us to help you further.

HTH.

Samer, PhD.

Dear Samer,

Thanks for your reply.

check the link below is an example for one approach to do the task that I've mentioned. (mapping low-level events to high level activities):

http://bpmcenter.org/wp-content/uploads/reports/2016/BPM-16-02.pdf

Also check the sample picture of the required dataset (taken from the paper of the previous link).

Dear Ghalia,

I'm trying to link your work with the area that I'm familiar with. This is the only help I can provide.

After obtaining the data, label this data based on your task. After that, perform feature extraction. Then feed the resulting data into a classification algorithm in order to build a predictive model under the evaluation of cross-validation, for instance.

NB: You could use a Ranker method to rank your features before classification. Moreover, association rules technique can help you to find the possible rules between your data features.

HTH.

Samer