Attention Please: What Transformer Models Really Learn for Process Prediction

TL;DR

This paper investigates whether attention scores in transformer models for process prediction can be used as explanations, proposing graph-based methods to interpret their decision-making and improve process modeling.

Contribution

It demonstrates that attention scores can serve as explanations in process prediction models and introduces graph-based methods for interpretability.

Findings

Attention scores can be used as explanations for predictions.

Graph-based explanation approaches improve interpretability.

Insights can enhance predictive process modeling and process mining.

Abstract

Predictive process monitoring aims to support the execution of a process during runtime with various predictions about the further evolution of a process instance. In the last years a plethora of deep learning architectures have been established as state-of-the-art for different prediction targets, among others the transformer architecture. The transformer architecture is equipped with a powerful attention mechanism, assigning attention scores to each input part that allows to prioritize most relevant information leading to more accurate and contextual output. However, deep learning models largely represent a black box, i.e., their reasoning or decision-making process cannot be understood in detail. This paper examines whether the attention scores of a transformer based next-activity prediction model can serve as an explanation for its decision-making. We find that attention scores in next-activity prediction models can serve as explainers and exploit this fact in two proposed graph-based explanation approaches. The gained insights could inspire future work on the improvement of predictive business process models as well as enabling a neural network based mining of process models from event logs.