Predictive Process Monitoring Methods: Which One Suits Me Best?

Francescomarino, Chiara Di; Ghidini, Chiara; Maggi, Fabrizio Maria; Milani, Fredrik

doi:10.1007/978-3-319-98648-7_27

Cited by 136 publications

(77 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…While some existing methods have tried to handle categorical and numeric attributes, some researchers have used text mining techniques in order to handle text attributes. Likewise, according to the paper [4], the researchers need to focus on the prediction target (results) from different perspectives. As the type of prediction, Numeric Prediction (Time Predictions, Cost Predictions), Categorical Prediction (Categorical Outcome, Risk Prediction), Activity Sequence Prediction are categorized.…”

Section: Predictive Process Monitoringmentioning

confidence: 99%

Attention Based LSTM with Multi Tasks Learning for Predictive Process Monitoring

2019

Proceedings of 2019 the 9th International Workshop on Computer Science and Engineering

View full text Add to dashboard Cite

Today, in the artificial intelligence research field, Deep Learning (DL) is one of the fastestgrowing techniques because of the power of learn ing features, that gives a higher level of abstraction of the raw attributes; and related research in Recurrent Neural Networks (RNN) and Long Short -Term Memo ry Networks (LSTM) have shown exemp lary results in neural machine translation, neural image caption generation, NLP and so on. For our research, in order to detect potential problems and to facilitate proactive management, we focus on predictive process monitoring (PPM) as domain are a, by predict ing business behaviour from h istorical event logs. Recent research works, LSTM networks have gained attention in PPM and have been proved that they can highly improve prediction accuracy in PPM. According to the literature, we have learned that PPM resembles to an early sequence classificat ion problem in NLP. And, recent trends in DL based NLP, attention mechanism is mostly embedded in neural networks. Inspired by these results, this paper proposes to firstly use Attention Based LSTM with Multi Tasks Learning for PPM.

show abstract

Section: Predictive Process Monitoringmentioning

confidence: 99%

Attention Based LSTM with Multi Tasks Learning for Predictive Process Monitoring

2019

Proceedings of 2019 the 9th International Workshop on Computer Science and Engineering

View full text Add to dashboard Cite

show abstract

“…Some predictive process monitoring approaches already use DL (Di Francescomarino et al 2018). Most of them strive for insights that help predict the next events during process execution (Evermann et al 2017a;Mehdiyev et al 2018;Pasquadibisceglie et al 2019;Tax et al 2017).…”

Section: Introductionmentioning

confidence: 99%

Machine Learning in Business Process Monitoring: A Comparison of Deep Learning and Classical Approaches Used for Outcome Prediction

et al. 2020

View full text Add to dashboard Cite

Predictive process monitoring aims at forecasting the behavior, performance, and outcomes of business processes at runtime. It helps identify problems before they occur and re-allocate resources before they are wasted. Although deep learning (DL) has yielded breakthroughs, most existing approaches build on classical machine learning (ML) techniques, particularly when it comes to outcome-oriented predictive process monitoring. This circumstance reflects a lack of understanding about which event log properties facilitate the use of DL techniques. To address this gap, the authors compared the performance of DL (i.e., simple feedforward deep neural networks and long short term memory networks) and ML techniques (i.e., random forests and support vector machines) based on five publicly available event logs. It could be observed that DL generally outperforms classical ML techniques. Moreover, three specific propositions could be inferred from further observations: First, the outperformance of DL techniques is particularly strong for logs with a high variant-to-instance ratio (i.e., many non-standard cases).Second, DL techniques perform more stably in case of imbalanced target variables, especially for logs with a high event-to-activity ratio (i.e., many loops in the control flow). Third, logs with a high activity-to-instance payload ratio (i.e., input data is predominantly generated at runtime) call for the application of long short term memory networks. Due to the purposive sampling of event logs and techniques, these findings also hold for logs outside this study.

show abstract

“…Predictive process monitoring predicts how an ongoing process instance (a.k.a. case) will unfold up to its completion [8,16,22]. If a potential problem is predicted, adaptation decisions are taken at run time to prevent or mitigate the predicted problem.…”

Section: Introductionmentioning

confidence: 99%

Proactive Process Adaptation Using Deep Learning Ensembles

Metzger

Neubauer

Bohn

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Proactive process adaptation can prevent and mitigate upcoming problems during process execution. Proactive adaptation decisions are based on predictions about how an ongoing process instance will unfold up to its completion. On the one hand, these predictions must have high accuracy, as, for instance, false negative predictions mean that necessary adaptations are missed. On the other hand, these predictions should be produced early during process execution, as this leaves more time for adaptations, which typically have non-negligible latencies. However, there is an important tradeoff between prediction accuracy and earliness. Later predictions typically have a higher accuracy, because more information about the ongoing process instance is available. To address this tradeoff, we use an ensemble of deep learning models that can produce predictions at arbitrary points during process execution and that provides reliability estimates for each prediction. We use these reliability estimates to dynamically determine the earliest prediction with sufficient accuracy, which is used as basis for proactive adaptation. Experimental results indicate that our dynamic approach may offer cost savings of 27% on average when compared to using a static prediction point.

show abstract

Predictive Process Monitoring Methods: Which One Suits Me Best?

Cited by 136 publications

References 67 publications

Attention Based LSTM with Multi Tasks Learning for Predictive Process Monitoring

Attention Based LSTM with Multi Tasks Learning for Predictive Process Monitoring

Machine Learning in Business Process Monitoring: A Comparison of Deep Learning and Classical Approaches Used for Outcome Prediction

Proactive Process Adaptation Using Deep Learning Ensembles

Contact Info

Product

Resources

About