# Mining Local Process Models

arXiv: https://arxiv.org/abs/1606.06066 · HTML: https://mlsanity.com/paper/1606.06066/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Jun 2016.

## Built on by (most cited first, 10 of 10)

- [Large Process Models: A Vision for Business Process Management in the Age of Generative AI](https://mlsanity.com/paper/2309.00900/) (2023)
- [Expert-driven Trace Clustering with Instance-level Constraints](https://mlsanity.com/paper/2110.06703/) (2021)
- [Grouping Local Process Models](https://mlsanity.com/paper/2311.03040/) (2023)
- [PSpan:Mining Frequent Subnets of Petri Nets](https://mlsanity.com/paper/2101.11972/) (2021)
- [Interactive Multi Interest Process Pattern Discovery](https://mlsanity.com/paper/2308.14475/) (2023)
- [Mining a Minimal Set of Behavioral Patterns using Incremental Evaluation](https://mlsanity.com/paper/2402.02921/) (2024)
- [An Invertible State Space for Process Trees](https://mlsanity.com/paper/2407.21468/) (2024)
- [On the Hybrid Nature of ABPMS Process Frames and its Implications on Automated Process Discovery](https://mlsanity.com/paper/2604.22455/) (2026)
- [TGCM: Topic-Guided Consistency Modeling for One-Step Disentanglement of Interleaved APT Technique Sequences](https://mlsanity.com/paper/2606.18651/) (2026)
- [Framework for Grouping Local Process Models](https://mlsanity.com/paper/2607.04856/) (2026)

## Abstract (arXiv)

In this paper we describe a method to discover frequent behavioral patterns in event logs. We express these patterns as local process models. Local process model mining can be positioned in-between process discovery and episode / sequential pattern mining. The technique presented in this paper is able to learn behavioral patterns involving sequential composition, concurrency, choice and loop, like in process mining. However, we do not look at start-to-end models, which distinguishes our approach from process discovery and creates a link to episode / sequential pattern mining. We propose an incremental procedure for building local process models capturing frequent patterns based on so-called process trees. We propose five quality dimensions and corresponding metrics for local process models, given an event log. We show monotonicity properties for some quality dimensions, enabling a speedup of local process model discovery through pruning. We demonstrate through a real life case study that mining local patterns allows us to get insights in processes where regular start-to-end process discovery techniques are only able to learn unstructured, flower-like, models.