DDTR: Diffusion Denoising Trace Recovery

2025-10-26

AI Summary

Existing trajectory recovery methods lack robustness under high noise when applied to stochastic process logs generated by nondeterministic data sources (e.g., noisy sensors or probabilistic AI models). This paper introduces, for the first time, diffusion denoising probabilistic models (DDPMs) into process mining, proposing a stochastic trajectory recovery method that jointly leverages explicit and implicit process knowledge. Specifically, it conditions the denoising process on trajectory distributions and incorporates process constraints to guide sampling, enabling reliable reconstruction of deterministic behavioral traces from probabilistic logs. The core innovation lies in a process-knowledge-driven diffusion framework, which significantly enhances noise robustness and recovery accuracy. Experiments across diverse noise scenarios demonstrate that the proposed method achieves an average 25% improvement in recovery accuracy over state-of-the-art approaches, along with markedly increased stability.

Technology Category

• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust
• Reasoning under Uncertainty: Stochastic Optimization
• Machine Learning: Probabilistic Circuits and Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

With recent technological advances, process logs, which were traditionally deterministic in nature, are being captured from non-deterministic sources, such as uncertain sensors or machine learning models (that predict activities using cameras). In the presence of stochastically-known logs, logs that contain probabilistic information, the need for stochastic trace recovery increases, to offer reliable means of understanding the processes that govern such systems. We design a novel deep learning approach for stochastic trace recovery, based on Diffusion Denoising Probabilistic Models (DDPM), which makes use of process knowledge (either implicitly by discovering a model or explicitly by injecting process knowledge in the training phase) to recover traces by denoising. We conduct an empirical evaluation demonstrating state-of-the-art performance with up to a 25% improvement over existing methods, along with increased robustness under high noise levels.

Problem

• Recovering stochastic traces from probabilistic process logs
• Denoising uncertain sensor and ML-generated activity data
• Improving reliability of process understanding in non-deterministic systems

Innovation

• Uses diffusion denoising probabilistic models for recovery
• Leverages process knowledge through model discovery or injection
• Demonstrates improved performance and robustness under noise