NegDis
Tool for the discovery of temporal-logic patterns as a binary supervised learning problem
The tool extracts Declare constraints starting from both positive and negative event log traces.
The NegDis tool - both compiled and source version - is available in the dist directory. Moreover, the data and the support code to run some experiments are also available in the repository.
The algorithms implemented in the tool, as well as the experiments available in the repository, are described in the paper "Process discovery on deviant traces and other stranger things" Federico Chesani, Chiara Di Francescomarino, Chiara Ghidini, Daniela Loreti, Fabrizio Maria Maggi, Paola Mello, Marco Montali, Sergio Tessaris. CoRR abs/2109.14883 (2021)
Additional information: The asset has been developed for the microproject "Discovering Temporal Logic Patterns as Binary Supervised Learning" that involves University of Bologna and Fondazione Bruno Kessler, as well as the Free University of Bolzano as external partner.