Predictive Business Process Monitoring with LSTM Neural Networks

Abstract

Predictive business process monitoring methods exploit logs of completed cases of a process in order to make predictions about running cases thereof. Existing methods in this space are tailor-made for specific prediction tasks. Moreover, their relative accuracy is highly sensitive to the dataset at hand, thus requiring users to engage in trial-and-error and tuning when applying them in a specific setting. This paper investigates Long Short-Term Memory (LSTM) neural networks as an approach to build consistently accurate models for a wide range of predictive process monitoring tasks. First, we show that LSTMs outperform existing techniques to predict the next event of a running case and its timestamp. Next, we show how to use models for predicting the next task in order to predict the full continuation of a running case. Finally, we apply the same approach to predict the remaining time, and show that this approach outperforms existing tailor-made methods.

Accepted at the International Conference on Advanced Information Systems Engineering (CAiSE) 2017

Cited by in corpus (46)

• Predictive Process Monitoring Methods: Which One Suits Me Best?
• A systematic literature review on state-of-the-art deep learning methods for process prediction
• Explainable Artificial Intelligence for Process Mining: A General Overview and Application of a Novel Local Explanation Approach for Predictive Process Monitoring
• Decay Replay Mining to Predict Next Process Events
• Temporal Stability in Predictive Process Monitoring
• Prescriptive Business Process Monitoring for Recommending Next Best Actions
• Building Interpretable Models for Business Process Prediction using Shared and Specialised Attention Mechanisms
• Process Outcome Prediction: CNN vs. LSTM (with Attention)
• Learning Uncertainty with Artificial Neural Networks for Improved Predictive Process Monitoring
• Learning Accurate Business Process Simulation Models from Event Logs via Automated Process Discovery and Deep Learning
• Classifying Process Instances Using Recurrent Neural Networks
• How do I update my model? On the resilience of Predictive Process Monitoring models to change
• ProcessTransformer: Predictive Business Process Monitoring with Transformer Network
• Counterfactual Explanations for Predictive Business Process Monitoring
• Tailoring Machine Learning for Process Mining
• An empirical comparison of deep-neural-network architectures for next activity prediction using context-enriched process event logs
• Text-Aware Predictive Monitoring of Business Processes
• An LSTM approach to Forecast Migration using Google Trends
• Automatically Reconciling the Trade-off between Prediction Accuracy and Earliness in Prescriptive Business Process Monitoring
• Local Post-Hoc Explanations for Predictive Process Monitoring in Manufacturing
• Event Log Sampling for Predictive Monitoring
• From Robotic Process Automation to Intelligent Process Automation: Emerging Trends
• Prescriptive Process Monitoring Under Resource Constraints: A Reinforcement Learning Approach
• Time Matters: Time-Aware LSTMs for Predictive Business Process Monitoring
• Process Model Forecasting Using Time Series Analysis of Event Sequence Data
• Generating the Traces You Need: A Conditional Generative Model for Process Mining Data
• Guiding the generation of counterfactual explanations through temporal background knowledge for Predictive Process Monitoring
• Long-term series forecasting with Query Selector -- efficient model of sparse attention
• Explainable Artificial Intelligence for Improved Modeling of Processes
• Can deep neural networks learn process model structure? An assessment framework and analysis
• Extending LIME for Business Process Automation
• Extending predictive process monitoring for collaborative processes
• Masking Neural Networks Using Reachability Graphs to Predict Process Events
• Prediction-based Resource Allocation using Bayesian Neural Networks and Minimum Cost and Maximum Flow Algorithm
• What Averages Do Not Tell -- Predicting Real Life Processes with Sequential Deep Learning
• An empirical investigation of different classifiers, encoding and ensemble schemes for next event prediction using business process event logs
• Mining Insights from Weakly-Structured Event Data
• Workload Prediction of Business Processes -- An Approach Based on Process Mining and Recurrent Neural Networks
• Creating Unbiased Public Benchmark Datasets with Data Leakage Prevention for Predictive Process Monitoring
• Specification-Driven Multi-Perspective Predictive Business Process Monitoring (Extended Version)
• Domain-independent generation and classification of behavior traces
• Predictive process mining by network of classifiers and clusterers: the PEDF model
• Application of Machine Learning in Early Recommendation of Cardiac Resynchronization Therapy
• Predictive Process Model Monitoring using Recurrent Neural Networks
• DeepProcess: Supporting business process execution using a MANN-based recommender system
• Leveraging Data Augmentation and Siamese Learning for Predictive Process Monitoring