PET Annotation Visualizer: A Tool to Visualize the Process Model Extraction from Text (PET) Dataset

The extraction of process models from natural language text is an essential task for process discovery. Several approaches have been developed in recent years to address this task. Still, one of the main limitations is the impossibility of visualizing the extracted process model and comparing it with gold standard data to assess similarity. In this paper, we present PET Visualizer, a visualization tool developed to graphically represent the process models of the document annotated in the PET dataset, the unique gold-standard dataset developed for process information extraction tasks. The main goal of PET Visualizer is to provide a way to visualize the dataset, making it easier to analyze and understand the process models. The tool supports several visualization options and can be easily integrated into existing workflows or extend the function implemented. During the system presentation, we demonstrate the effectiveness of the tool and its potential impact in improving process extraction from text tasks. The tool is freely available as a Python package at pypi.org/project/PETAnnotationVisualizer.