Exploring LLM Features in Predictive Process Monitoring for Small-Scale Event-Logs

2026-01-16

AI Summary

This work addresses the challenge of predictive process monitoring in data-scarce scenarios, where traditional methods suffer from limited performance due to the availability of only a small number of traces—on the order of one hundred. To overcome this limitation, the authors propose a novel prediction framework leveraging large language models (LLMs) without requiring fine-tuning. By means of carefully designed prompt engineering, the approach harnesses the LLM’s rich prior knowledge and its capacity to infer semantic relationships among traces, enabling high-order reasoning for accurate prediction of key performance indicators such as total cycle time and activity occurrence. Experimental results on three real-world event logs demonstrate that the proposed method outperforms existing benchmarks using merely 100 traces, significantly enhancing predictive accuracy and generalization under few-shot conditions, thereby highlighting the potential of LLMs in data-scarce process mining tasks.

Technology Category

• Data Mining & Knowledge Management: Intelligent Query Processing
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Predictive Process Monitoring is a branch of process mining that aims to predict the outcome of an ongoing process. Recently, it leveraged machine-and-deep learning architectures. In this paper, we extend our prior LLM-based Predictive Process Monitoring framework, which was initially focused on total time prediction via prompting. The extension consists of comprehensively evaluating its generality, semantic leverage, and reasoning mechanisms, also across multiple Key Performance Indicators. Empirical evaluations conducted on three distinct event logs and across the Key Performance Indicators of Total Time and Activity Occurrence prediction indicate that, in data-scarce settings with only 100 traces, the LLM surpasses the benchmark methods. Furthermore, the experiments also show that the LLM exploits both its embodied prior knowledge and the internal correlations among training traces. Finally, we examine the reasoning strategies employed by the model, demonstrating that the LLM does not merely replicate existing predictive methods but performs higher-order reasoning to generate the predictions.

Problem

• Predictive Process Monitoring
• Small-Scale Event Logs
• LLM
• Process Mining
• Key Performance Indicators

Innovation

• Large Language Models
• Predictive Process Monitoring
• Small-Scale Event Logs
• Higher-Order Reasoning
• Process Mining