Building interpretable models for business process prediction using shared and specialised attention mechanisms

Wickramanayake, Bemali; He, Zhipeng; Ouyang, Chun; Moreira, Catarina; Xu, Yue; Sindhgatta, Renuka

doi:10.1016/j.knosys.2022.108773

Cited by 24 publications

(23 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Bukhsh et al [11] proposed the Process Transformer model, that is, to modify the structure of the Transformer network according to specific process prediction tasks to achieve ideal prediction results. Similarly, Wickramanayake et al [28] proposed two types of attention for the prediction task of future activities: the event-level attention to capturing the impact of specific events on the prediction task and the attribute-level attention to reveal which attributes of events affect the prediction task. And they designed two different attention models, the shared and specialized attention-based models.…”

Section: B Prediction Based On Deep Learningmentioning

confidence: 99%

MTLFormer: Multi-Task Learning Guided Transformer Network for Business Process Prediction

Jiaojiao

Huang

et al. 2023

IEEE Access

View full text Add to dashboard Cite

The predictive business process monitoring mainly focuses on the performance prediction of business process execution, i.e., predicting the next activity, the execution time of the next activity, and the remaining time, respectively, for an ongoing process instance based on the knowledge gained from historical event logs. Although there is a specific relationship between these three tasks, recent research has focused on training separate prediction models for each task, resulting in high costs and time complexity. Additionally, existing technologies are limited in their ability to capture long-distance dependent features in process instances, further impeding prediction performance. To address these issues, this paper proposes the MTLFormer approach, which leverages the self-attention mechanism of the Transformer network and conducts multi-task parallel training through shared feature representation obtained from different tasks. Our approach reduces the time complexity of model training while simultaneously improving prediction performance. We extensively evaluate our approach on four real-life event logs, demonstrating its capability to achieve multi-task online real-time prediction and effectively improve prediction performance.

show abstract

Section: B Prediction Based On Deep Learningmentioning

confidence: 99%

MTLFormer: Multi-Task Learning Guided Transformer Network for Business Process Prediction

Jiaojiao

Huang

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The framework proposed in this paper specializes the use of Shapley values to the problem of providing explanations for predictive analytics. Furthermore, the use of SHAP enabled us to provide explanations independently from the leveraged predictive model (model agnostic), while model specific approaches are specifically designed for certain model types; in particular, some approaches [38,39,40,41,42,43] can provide explanations only for neural network models. Other works [16,40] use attention mechanisms, which also have the limitation that is linked to the lack of consensus that attention weights are always correlated to feature importance.…”

Section: Explanations In the Bpm Fieldmentioning

confidence: 99%

An Explainable Decision Support System for Predictive Process Analytics

Galanti¹,

Leoni²,

Monaro³

et al. 2022

Preprint

View full text Add to dashboard Cite

Predictive Process Analytics is becoming an essential aid for organizations, providing online operational support of their processes. However, process stakeholders need to be provided with an explanation of the reasons why a given process execution is predicted to behave in a certain way. Otherwise, they will be unlikely to trust the predictive monitoring technology and, hence, adopt it. This paper proposes a predictive analytics framework that is also equipped with explanation capabilities based on the game theory of Shapley Values. The framework has been implemented in the IBM Process Mining suite and commercialized for business users. The framework has been tested on real-life event data to assess the quality of the predictions and the corresponding evaluations. In particular, a user evaluation has been performed in order to understand if the explanations provided by the system were intelligible to process stakeholders.

show abstract

“…Attention mechanism is a resource allocation scheme, which can strengthen the ability of the model to focus on more critical information [35][36][37][38][39][40][41][42]. The principle of attention mechanism used in the paper is shown in Fig.…”

Section: Attention Mechanism In the Clstm Modelmentioning

confidence: 99%

A multi-task prediction method for acid concentration based on attention-CLSTM

Zhu

Liu

Cao

et al. 2022

Preprint

View full text Add to dashboard Cite

The accurate measurement of acid concentration, including hydrogen chloride (HCl) and ferrous chloride concentrations (FeCl2), is a critical part of ensuring the quality of strip steel pickling. In this study, a multi-task attention convolutional long short-term memory (MACL) neural network model was proposed to predict hydrogen ion and ferrous ion concentrations simultaneously. Firstly, in order to extract significant information from the input sequence, an attention mechanism was added to the model to calculate the correlation between the input features and the acid concentration at each time step. Next, multi-task learning explores the connections between the two tasks and exploits hidden constraints to improve prediction accuracy. Finally, deep features were extracted through convolutional long short-term memory (CLSTM) neural network, and the acid concentration was predicted. The proposed MACL model was compared with other popular prediction models. The experimental results show that proposed MACL model generally outperforms other models, indicating that proposed model has excellent predictive performance and effectiveness.

show abstract

Building interpretable models for business process prediction using shared and specialised attention mechanisms

Cited by 24 publications

References 15 publications

MTLFormer: Multi-Task Learning Guided Transformer Network for Business Process Prediction

MTLFormer: Multi-Task Learning Guided Transformer Network for Business Process Prediction

An Explainable Decision Support System for Predictive Process Analytics

A multi-task prediction method for acid concentration based on attention-CLSTM

Contact Info

Product

Resources

About