Evaluating Stability of Post-hoc Explanations for Business Process Predictions

Velmurugan, Mythreyi; Ouyang, Chun; Moreira, Catarina; Sindhgatta, Renuka

doi:10.1007/978-3-030-91431-8_4

Cited by 8 publications

(8 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Also [10] reports on the use of post-hoc, SHAP explanations for both the LSTM and the CatBoost method. [28] compares and evaluates explanations of process predictions yielded by different post-hoc frameworks (e.g., LIME and SHAP). [29] leverages post-hoc explainers to understand why a PPM model provides wrong predictions, eventually improving its accuracy.…”

Section: Explainability In Ppmmentioning

confidence: 99%

JARVIS: Joining Adversarial Training With Vision Transformers in Next-Activity Prediction

Pasquadibisceglie,

Appice,

Castellano

et al. 2024

IEEE Trans. Serv. Comput.

View full text Add to dashboard Cite

In this paper, we propose a novel predictive process monitoring approach, named JARVIS, that is designed to achieve a balance between accuracy and explainability in the task of next-activity prediction. To this aim, JARVIS represents different process executions (traces) as patches of an image and uses this patch-based representation within a multi-view learning scheme combined with Vision Transformers (ViTs). Using multi-view learning we guarantee good accuracy by leveraging the variety of information recorded in event logs as different patches of an image. The use of ViTs enables the integration of explainable elements directly into the framework of a predictive process model trained to forecast the next trace activity from the completed events in a running trace by utilizing self-attention modules that give paired attention values between two picture patches. Attention modules disclose explainable information concerning views of the business process and events of the trace that influenced the prediction. In addition, we explore the effect of ViT adversarial training to mitigate overfitting and improve the accuracy and robustness of predictive process monitoring. Experiments with various benchmark event logs prove the accuracy of JARVIS compared to several current state-of-the-art methods and draw insights from explanations recovered through the attention modules.

show abstract

Section: Explainability In Ppmmentioning

confidence: 99%

JARVIS: Joining Adversarial Training With Vision Transformers in Next-Activity Prediction

Pasquadibisceglie,

Appice,

Castellano

et al. 2024

IEEE Trans. Serv. Comput.

View full text Add to dashboard Cite

show abstract

“…Particularly, it is necessary to evaluate the ability of an XAI method to reflect the knowledge learned by an ML model. With the increasing interest in applying XAI methods in PPM, a few proposals [3]- [5] were made to evaluate the explanations created for PPM results. This article proposes an approach for evaluating XAI methods with respect to their ability to transfer data facts learned by an ML model about PPM data.…”

Section: A Problem Statementmentioning

confidence: 99%

“…An evaluation method has a target characteristic against which the performance of an XAI method is evaluated. For example, there exist methods for evaluating an explanation for its stability [5], [20], understandability [3], robustness [21], [22] and fidelity [4]. Our proposed approach evaluates an XAI method with respect to the degree of its consistency with the underlying data.…”

Section: B Explainable Artificial Intelligence Applicationmentioning

confidence: 99%

Why Should I Trust Your Explanation? An Evaluation Approach for XAI Methods Applied to Predictive Process Monitoring Results

Elkhawaga,

Elzeki,

Abu-Elkheir

et al. 2024

IEEE Trans. Artif. Intell.

View full text Add to dashboard Cite

“…So there is no consistent way of finding this minimum counterfactual. Although there are some metrics that are used for feature attribution XAI algorithms, such as fidelity [16] and stability [17], there is no standardised way of evaluating XAI algorithms in general, which increases the complexity and difficulty of developing a benchmark evaluation for counterfactuals.…”

Section: The Problem Of Validation Of Counterfactual Explanationsmentioning

confidence: 99%

Benchmark Evaluation of Counterfactual Algorithms for XAI: From a White Box to a Black Box

Chou¹,

Hsieh²,

Moreira³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Counterfactual explanations have recently been brought to light as a potentially crucial response to obtaining human-understandable explanations from predictive models in Explainable Artificial Intelligence (XAI). This counterfactual explanation claims to be capable of identifying the smallest feature change capable of changing a prediction. It produces an explanation that is more causally informative than factual explanations in terms of expanding human knowledge of decision making. Despite the fact that various counterfactual algorithms have been proposed, the state of the art research still lacks on standardised protocols to evaluate the quality of counterfactual explanations.In this work, we conducted a benchmark evaluation across different model agnostic counterfactual algorithms in the literature (DiCE, WatcherCF, prototype, unjustifiedCF), and we investigated the counterfactual generation process on different types of machine learning models ranging from a white box (decision tree) to a grey-box (random forest) and a black box (neural network). We evaluated the different counterfactual algorithms using several metrics including proximity, interpretability and functionality for five datasets.The main findings of this work are the following: (1) without guaranteeing plausibility in the counterfactual generation process, one cannot have meaningful evaluation results. This means that all explainable counterfactual algorithms that do not take into consideration plausibility in their internal mechanisms cannot be evaluated with the current state of the art evaluation metrics; (2) the counterfactual generated are not impacted by the different types of machine learning models; (3) DiCE was the only tested algorithm that was able to generate actionable and plausible counterfactuals, because it provides mechanisms to constraint features; (4) WatcherCF and UnjustifiedCF are limited to continuous variables and can not deal with categorical data.

show abstract

Evaluating Stability of Post-hoc Explanations for Business Process Predictions

Cited by 8 publications

References 9 publications

JARVIS: Joining Adversarial Training With Vision Transformers in Next-Activity Prediction

JARVIS: Joining Adversarial Training With Vision Transformers in Next-Activity Prediction

Why Should I Trust Your Explanation? An Evaluation Approach for XAI Methods Applied to Predictive Process Monitoring Results

Benchmark Evaluation of Counterfactual Algorithms for XAI: From a White Box to a Black Box

Contact Info

Product

Resources

About