Validation of AI-based Information Systems for Sensitive Use Cases: Using an XAI Approach in Pharmaceutical Engineering

Polzer, Anna; Fleiß, Jürgen; Ebner, Thomas; Kainz, Philipp; Koeth, Christoph; Thalmann, Stefan

doi:10.24251/hicss.2022.186

Cited by 6 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Researchers and practitioners are interested in XAI business models to help them explore data relationships, improve AI methods, justify AI decisions, and control XAI technologies while simultaneously meeting user needs (Adadi & Berrada, 2018;Meske et al, 2022;Thiebes et al, 2021). In contrast, many other scientists have focused on XAI algorithms and proposed artifacts to increase unbiased AI decision-making (Xie et al, 2022) or to understand the behavior of an AI system (Polzer et al, 2022). To benefit from such solutions, users and interested stakeholders such as managers, data scientists, and AI developers must determine which XAI solution best fits their requirements.…”

Section: Discussion and A Decision Support Frameworkmentioning

confidence: 99%

Decision support for efficient XAI services - A morphological analysis, business model archetypes, and a decision tree

et al. 2022

View full text Add to dashboard Cite

The black-box nature of Artificial Intelligence (AI) models and their associated explainability limitations create a major adoption barrier. Explainable Artificial Intelligence (XAI) aims to make AI models more transparent to address this challenge. Researchers and practitioners apply XAI services to explore relationships in data, improve AI methods, justify AI decisions, and control AI technologies with the goals to improve knowledge about AI and address user needs. The market volume of XAI services has grown significantly. As a result, trustworthiness, reliability, transferability, fairness, and accessibility are required capabilities of XAI for a range of relevant stakeholders, including managers, regulators, users of XAI models, developers, and consumers. We contribute to theory and practice by deducing XAI archetypes and developing a user-centric decision support framework to identify the XAI services most suitable for the requirements of relevant stakeholders. Our decision tree is founded on a literature-based morphological box and a classification of real-world XAI services. Finally, we discussed archetypical business models of XAI services and exemplary use cases.

show abstract

Section: Discussion and A Decision Support Frameworkmentioning

confidence: 99%

Decision support for efficient XAI services - A morphological analysis, business model archetypes, and a decision tree

et al. 2022

View full text Add to dashboard Cite

show abstract

“…It is possible to associate biological effects with physicochemical effects and to derive accurate and appropriate models according to this relationship. Ultimately, XAI aims to reveal what is done, how it is done, and related information in drug discovery (Holzinger et al, 2022; Polzer et al, 2022). Given the importance of explainability, XAI is emerging, a collection of AI methods focused on generating outputs and recommendations that human experts can understand and interpret.…”

Section: Ai and Xai In Drug Discoverymentioning

confidence: 99%

Explainability and white box in drug discovery

2023

View full text Add to dashboard Cite

Recently, artificial intelligence (AI) techniques have been increasingly used to overcome the challenges in drug discovery. Although traditional AI techniques generally have high accuracy rates, there may be difficulties in explaining the decision process and patterns. This can create difficulties in understanding and making sense of the outputs of algorithms used in drug discovery. Therefore, using explainable AI (XAI) techniques, the causes and consequences of the decision process are better understood. This can help further improve the drug discovery process and make the right decisions. To address this issue, Explainable Artificial Intelligence (XAI) emerged as a process and method that securely captures the results and outputs of machine learning (ML) and deep learning (DL) algorithms. Using techniques such as SHAP (SHApley Additive ExPlanations) and LIME (Locally Interpretable Model-Independent Explanations) has made the drug targeting phase clearer and more understandable. XAI methods are expected to reduce time and cost in future computational drug discovery studies.This review provides a comprehensive overview of XAI-based drug discovery and development prediction. XAI mechanisms to increase confidence in AI and modeling methods. The limitations and future directions of XAI in drug discovery are also discussed.

show abstract

“…Automation bias describes the tendency of people to thoughtlessly accept an automated decision or recommendation. Thus far, automation bias and its negative outcomes have primarily been investigated in aviation contexts (e.g., Mosier and Skitka, 1999;Davis et al, 2020) and medical contexts (e.g., Goddard et al, 2012;Lyell et al, 2018), but have also been found in the military domain and in process control (Bahner et al, 2008;Parasuraman and Manzey, 2010) as well as in quality control (Kloker et al, 2022). However, automation bias can occur in every work field that includes human-systeminteraction (Goddard et al, 2012).…”

Section: Human Information Processing and Automation Biasmentioning

confidence: 99%

Check the box! How to deal with automation bias in AI-based personnel selection

et al. 2023

Self Cite

View full text Add to dashboard Cite

Artificial Intelligence (AI) as decision support for personnel preselection, e.g., in the form of a dashboard, promises a more effective and fairer selection process. However, AI-based decision support systems might prompt decision makers to thoughtlessly accept the system’s recommendation. As this so-called automation bias contradicts ethical and legal requirements of human oversight for the use of AI-based recommendations in personnel preselection, the present study investigates strategies to reduce automation bias and increase decision quality. Based on the Elaboration Likelihood Model, we assume that instructing decision makers about the possibility of system errors and their responsibility for the decision, as well as providing an appropriate level of data aggregation should encourage decision makers to process information systematically instead of heuristically. We conducted a 3 (general information, information about system errors, information about responsibility) x 2 (low vs. high aggregated data) experiment to investigate which strategy can reduce automation bias and enhance decision quality. We found that less automation bias in terms of higher scores on verification intensity indicators correlated with higher objective decision quality, i.e., more suitable applicants selected. Decision makers who received information about system errors scored higher on verification intensity indicators and rated subjective decision quality higher, but decision makers who were informed about their responsibility, unexpectedly, did not. Regarding aggregation level of data, decision makers of the highly aggregated data group spent less time on the level of the dashboard where highly aggregated data were presented. Our results show that it is important to inform decision makers who interact with AI-based decision-support systems about potential system errors and provide them with less aggregated data to reduce automation bias and enhance decision quality.

show abstract

Validation of AI-based Information Systems for Sensitive Use Cases: Using an XAI Approach in Pharmaceutical Engineering

Cited by 6 publications

References 15 publications

Decision support for efficient XAI services - A morphological analysis, business model archetypes, and a decision tree

Decision support for efficient XAI services - A morphological analysis, business model archetypes, and a decision tree

Explainability and white box in drug discovery

Check the box! How to deal with automation bias in AI-based personnel selection

Contact Info

Product

Resources

About