Text-Aware Predictive Monitoring of Business Processes

  title={Text-Aware Predictive Monitoring of Business Processes},
  author={Marco Pegoraro and Merih Seran Uysal and David Benedikt Georgi and Wil M.P. van der Aalst},
The real-time prediction of business processes using historical event data is an important capability of modern business process monitoring systems. Existing process prediction methods are able to also exploit the data perspective of recorded events, in addition to the control-flow perspective. However, while well-structured numerical or categorical attributes are considered in many prediction techniques, almost no technique is able to utilize text documents written in natural language, which… 

Figures and Tables from this paper

Predictive Compliance Monitoring in Process-Aware Information Systems: State of the Art, Functionalities, Research Directions
This work analyzes existing literature from compliance and SLA monitoring as well as predictive process monitoring and provides an updated framework of compliance monitoring functionalities and open challenges and research directions for predictive compliance and process monitoring are elaborated.
Event Log Sampling for Predictive Monitoring
This paper proposes an instance selection procedure that allows sampling training process instances for prediction models and shows that the sampling method allows for a significant increase of training speed for next activity prediction methods while maintaining reliable levels of prediction accuracy.


LSTM networks for data-aware remaining time prediction of business process instances
This paper proposes an approach based on deep Recurrent Neural Networks (specifically LSTMs) that is able to exploit arbitrary information associated to single events in order to produce an as-accurate-as-possible prediction of the completion time of running instances.
Latent Dirichlet Allocation
Predicting performances in business processes using deep neural networks
Process Mining for Python (PM4Py): Bridging the Gap Between Process- and Data Science
A novel process mining library that aims to bridge the gap between commercial and open-source process mining tools, providing integration with state-of-the-art data science libraries, e.g., pandas, numpy, scipy and scikit-learn is presented.
An interdisciplinary comparison of sequence modeling methods for next-element prediction
An interdisciplinary experimental evaluation that compares sequence modeling methods on the task of next-element prediction on four real-life sequence datasets indicates that machine learning methods, which generally do not aim at model interpretability, tend to outperform methods from the process mining and grammar inference fields in terms of accuracy.
Predictive Business Process Monitoring with Structured and Unstructured Data
This paper presents a predictive process monitoring framework that combines text mining with sequence classification techniques so as to handle both structured and unstructured event payloads.
A Deep Learning Approach for Predicting Process Behaviour at Runtime
This paper describes an initial application of deep learning with recurrent neural networks to the problem of predicting the next process event, which is both a novel method in process prediction, and also a novel application for deep learning methods.
Predictive Business Process Monitoring with LSTM Neural Networks
This paper investigates Long Short-Term Memory neural networks as an approach to build consistently accurate models for a wide range of predictive process monitoring tasks and shows that LSTMs outperform existing techniques to predict the next event of a running case and its timestamp.
TensorFlow: A system for large-scale machine learning
The TensorFlow dataflow model is described and the compelling performance that Tensor Flow achieves for several real-world applications is demonstrated.
Time and activity sequence prediction of business process instances
A method, assuming process stationarity, is proposed which achieves state-of-the-art performances and two other methods which are able to make predictions even with non-stationary processes are described, able to predict the full sequence of activities that a running case is going to take.