What is it about?

This paper proposes for the first time a temporal declarative process mining algorithm exploiting support and confidence heuristics and a first adoption of the notion of logical entailment for reducing the size of the temporal mined model.

Featured Image

Why is it important?

This paper proposes for the first time a temporal declarative process mining algorithm exploiting support and confidence heuristics and an ad-hoc indexed relational database for outperforming existing temporal declarative data mining algorithms.

Perspectives

While the previous declarative mining approaches were mainly based on a priori pattern mining algorithms and on the instantiation of all of the possible declarative patterns for then testing their presence in the temporal data, this algorithm changed the approach for this drastically: we exploit specific heuristics to determine whether the data will support the declarative temporal clause and, only if not discarded, we are then going to test the clause. Compared to our previous work, we also reduce the memory footprint by leveraging the indexed data as much as possible without generating intermediate results.

Dr Giacomo Bergami
Newcastle University

Read the Original

This page is a summary of: Enhancing Declarative Temporal Model Mining in Relational Databases: A Preliminary Study, May 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3589462.3589491.
You can read the full text:

Read

Contributors

The following have contributed to this page