Designing User-Centric Explanations for Medical Imaging with Informed Machine Learning

2023 | DESRIST | Citations: 0

Authors: Oberste, Luis; Rüffer, Florian; Aydingül, Okan; Rink, Johann; Heinzl, Armin

Abstract: A flawed algorithm released in clinical practice can cause unintended harm to pa ... Expand

Abstract: A flawed algorithm released in clinical practice can cause unintended harm to patient health. Risks, regulation, responsibility, and ethics shape the demand of clinical users to understand and rely on the outputs made by artificial intelligence. Explainable artificial intelligence (XAI) offers methods to render a model’s behavior understandable from different perspectives. Extant XAI, however, is mainly data-driven and designed to meet developers’ demands to correct models rather than clinical users’ expectations to reflect clinically relevant information. To this end, informed machine learning (IML) utilizes prior knowledge jointly with data to generate predictions, a promising paradigm to enrich XAI with medical knowledge. To explore how IML can be used to generate explanations that are congruent to clinical users’ demands and useful to medical decision-making, we conduct Action Design Research (ADR) in collaboration with a team of radiologists. We propose an IML-based XAI system for clinically relevant explanations of diagnostic imaging predictions. With the help of ADR, we reduce the gap between implementation and user evaluation and demonstrate the effectiveness of the system in a real-world application with clinicians. While we develop design principles of using IML for user-centric XAI in diagnostic imaging, the study demonstrates that an IML-based design adequately reflects clinicians’ conceptions. In this way, IML inspires greater understandability and trustworthiness of AI-enabled diagnostic imaging. Collapse

Topics: explainable artificial intelligence artificial intelligence usability knowledge base decision making

Methods: design artifact design principle qualitative interview design science action design research

One Save per Day: How mobile technology can support individuals to adopt pro-environmental behaviors

2023 | HICSS | Citations: 0

Authors: Ochmann, J.; Lehrer, C.

Abstract: The pressing issue of climate change requires humanity to reduce its ecological ... Expand

Abstract: The pressing issue of climate change requires humanity to reduce its ecological footprint drastically. While policymakers and companies must ensure the availability of green options, individuals are requested to contribute to the reduction of carbon emissions substantially. However, even when individuals recognize the need for pro-environmental behaviors, they often have difficulty meeting their expectations. Mobile technology for sustainability has the potential to support them in overcoming this issue by providing the decisive impetus for environmentally friendly behavior. Drawing upon the affordance perspective, we conducted a longitudinal qualitative study with users of a mobile app that encourages individuals to take daily sustainable actions. We present the affordance strands made possible by the app's features and how they lead to environmentally-friendly behaviors. We could observe behavior change with the app's features. We also identify enablers and obstacles to affordance actualization. Our study contributes to Green IS research at the individual level and provides practical implications for mobile technology providers. Collapse

Topics: mobile application affordance green IS usability user experience

Methods: qualitative interview survey longitudinal research survey design qualitative coding

Theories: theory of affordance

SMART AGRICULTURE: SUITABLE CROPS FOR AUTONOMOUS SELECTIVE HARVESTING

2023 | European Conference On Information Systems | Citations: 0

Authors: Ockenga, Tim; Hirschmeier, Stefan; Mazur, Philipp; Schoder, Detlef

Abstract: Agriculture is one of the most important economic sectors and essential for our ... Expand

Abstract: Agriculture is one of the most important economic sectors and essential for our daily life. Especially in relation to selectively harvested crops, agriculture is majorly affected by increasing wage demands and workforce shortages. While other industries tackle this issue with increasing automation, the task of harvesting fruit and vegetables is still largely fulfilled manually. Following an overview of selective harvesting robot architecture, this paper addresses the question of which crops are ideally suited for AI-assisted selective harvesting. For this purpose, the state of development of robots for specific crops is evaluated based on performance measurements, and crop characteristics are determined through interviews with domain experts. We found that cauliflower and broccoli have limited suitability, sweet peppers, apple and cucumber have good suitability, and kiwi, strawberry and tomato are most suitable. Collapse

Topics: robotic artificial intelligence innovation management Google Search

Methods: qualitative interview literature study structured literature research qualitative content analysis

Smart Agriculture: Technical Components in Selective Harvesting

2023 | Americas Conference on Information Systems | Citations: 0

Authors: Ockenga, Tim Alvaro; Hirschmeier, Stefan; Mazur, Philipp; Schoder, Detlef

Abstract: Harvesting presents one of the most challenging tasks in the agricultural field ... Expand

Abstract: Harvesting presents one of the most challenging tasks in the agricultural field. While other areas within and outside the agricultural sector have moved to automate labor-intensive tasks, the selective harvesting of fruit still involves painstaking manual labor. Although progress has been made in the development of selective harvesting robots in agriculture, the proposed solutions still do not meet the requirements for commercialization and need further improvement from a technical point of view. To address the need to assess the current developments in technical components, we conduct a systematic literature review on recent publications in the field and depict our findings along three main technical components, namely vision system, manipulator & end-effector and robot control. The obtained findings show that a significant body of literature focuses on the vision system, while the results on manipulator and end effector and especially on robot control are limited. Collapse

Topics: robotic database system

Methods: literature study structured literature research literature sample qualitative content analysis

The Price of Belonging: Price Setting in the Market for Champagne Grapes

2023 | Organization Science | Citations: 0

Authors: Ody-Brasier, Amandine

Abstract: Various studies show that social relations between exchange partners affect pric ... Expand

Abstract: Various studies show that social relations between exchange partners affect price setting behaviors. Although scholars expect a broad range of buyer–seller relations to shape price—and different types of relations to have different effects—we have surprisingly little quantitative evidence for these views. This is in part because extant empirical work operationalizes relations rather narrowly, focusing on repeated exchange relations. Drawing on existing theory and fieldwork in the market for Champagne grapes, this paper investigates how community relations and repeated exchange shape price setting behavior. I use unique transaction-level price data to document the effects of these two types of buyer–seller relations. I find that community relations reduce deviations from local pricing norms such that, for an average transaction, grape sellers charge about €1,500 less when the buyer belongs to their village or school district. I draw on rich qualitative data to illustrate what is driving these effects. Interviews reveal not only that social relations infuse in the exchange a logic of action at odds with the logic of markets, but also, importantly, that there is some heterogeneity within the embedded logic of exchange. In particular, community relations appear to be more constraining than repeated exchange for price setters. Supplemental Material: The online appendix is available at https://doi.org/10.1287/orsc.2021.1564. Collapse

Topics: price management accounting willingness to pay geographic information system IT recruiting

Methods: field study qualitative interview Wald test longitudinal research archival research

Theories: transaction cost economics

Collaborative Work Practices for Management Education: Using Collaboration Engineering to Design a Reusable and Scalable Collaborative Learning Instructional Design

2023 | HICSS | Citations: 0

Authors: Oeste-Reiß, S.; Söllner, M.; Leimeister, J.

Abstract: Pandemics like COVID-19 highlight the needs and pitfalls of inclusive and equit ... Expand

Abstract: Pandemics like COVID-19 highlight the needs and pitfalls of inclusive and equitable education in a digital society. IT-based instructional designs are needed to increase learners' expertise, and to develop higherorder thinking skills. Instructional designs for collaborative learning (CL) seem to be a promising solution. However, they are mostly suitable for face-toface and not for distance teaching. The core problem that impedes their reusability and scalability is a 'collaboration problem' for which collaboration engineering (CE) provides guidance. Therefore, we deploy a design science research study and contribute to CL and CE literature. We develop requirements and provide the design of an IT-based collaborative work practice fostering CL. We provide empirical evidence with an online experiment in a large-scale lecture with undergraduate business information students. This reveals that groups of learners who followed our CL experience achieve higher levels of expertise than those who followed a traditional ad hoc CL experience. Collapse

Topics: online learning system team productivity productivity collaborative work spreadsheet

Methods: experiment design science experimental task case study nonparametric test

Transferring Well-Performing Collaborative Work Practices with Parameterized Templates and Guidebooks: Empowering Subject Matter Experts for an Adaptation to Slightly Different Contexts

2023 | HICSS | Citations: 0

Authors: Oeste-Reiß, S.; Söllner, M.; Leimeister, J.

Abstract: Collaborative work practices (CWPs) package facilitation expertise and have the ... Expand

Abstract: Collaborative work practices (CWPs) package facilitation expertise and have the potential to increase team productivity up to 90%. Collaboration engineers develop CWPs and deploy them to practitioners that execute them. These CWPs, however, are typically customized to conditions of a specific use case. This creates the challenge that changing use case conditions or even small variations across contexts, hinder well-performing CWPs of being applied more often to create a long-term value. Practitioners fail to adapt existing CWPs due to missing collaboration expertise and adaptation guidelines. To address this challenge in collaboration engineering literature, we introduce a) the Subject Matter Expert role; b) the 'CWP Adaptation Approach' that formalizes the transfer of CWPs to different contexts with parameterized Templates and Guidebooks. To show a first proof-of-concept, we further inductively generalize from an exemplarily use case with a well-performing CWP in the educational domain. Collapse

Topics: value creation collaborative work human resource management team productivity

Methods: design artifact design methodology

Data Analytics Platforms: Value Propositions and Adoption Challenges for Small Hospitality Businesses

2023 | HICSS | Citations: 0

Authors: Ofe, Hosea; De Reuver, Mark; Nederstigt, Bouke; Janssen, Marijn

Abstract: Managers increasingly seek ways to explore insights from data for business impro ... Expand

Abstract: Managers increasingly seek ways to explore insights from data for business improvements and innovation. Data Analytics (DA) platforms hold promise for businesses, especially small businesses that cannot afford tailor-made proprietary analytics services. DA platforms offer generic analytics features to a pool of businesses, saving costs and enabling benchmarking. This paper explores value propositions and adoption challenges for small businesses regarding DA. The paper offers practical insights from designing and launching a DA platform targeting small businesses in the hospitality sector. The findings of our paper show that data analytics is potentially valuable for small businesses through insights into market and customer trends. Small businesses can leverage such insights to refine their offerings. Trust and privacy concerns in sharing data are key challenges holding back adoption. We proposed measures, especially privacy-preserving technologies, to mitigate the risk of tracing a specific enterprise's data shared on the DA platform. These measures assure businesses that data shared or analyzed through the DA platform is not used to harm their competitive advantage. Collapse

Topics: value creation technology adoption analytical information system digital dashboard usage intention

Methods: qualitative interview case study personal interview

Towards an Architecture for Data Monetization as a Service

2023 | Americas Conference on Information Systems | Citations: 0

Authors: Ofulue, Joan; Benyoucef, Morad

Abstract: Data monetization as a service (DMaaS) is a new term in both academia and pract ... Expand

Abstract: Data monetization as a service (DMaaS) is a new term in both academia and practice. DMaaS seeks to marry data monetization with servitization and has become important with the advent of technologies such as big data, cloud computing and blockchain. As organizations seek to stand up infrastructure for data monetization, an architecture for DMaaS provides them with a starting point to develop data monetization platforms. This is an area that has been largely unexplored. Therefore, in this paper we leverage the service dominant logic perspective and research on data related services architecture to develop a tentative design for DMaaS. This study is among the first that explore architecture for DMaaS, thus identifying platform layers and services as a base for further architecture development and refinement. Collapse

Topics: big data cloud computing business-to-business database system database as a service

Methods: structured literature research qualitative interview literature study design process design requirement

Theories: service-dominant logic

Using Multi-Factor Authentication for Online Account Security: Examining the Influence of Anticipated Regret

2023 | Information Systems Frontiers | Citations: 0

Authors: Ogbanufe, Obi M.; Baham, Corey

Abstract: Authentication plays an important role in securing our systems but is threatened ... Expand

Abstract: Authentication plays an important role in securing our systems but is threatened by increasingly sophisticated account hacking and account take over. Several security services have been developed, including multi-factor authentication (MFA) designed to cope with online account authentication. It remains unknown how users perceive and evaluate secure authentication for online account threats and consequently use it to protect their online account. Drawing on the Protection Motivation Theory (PMT) and the literature on anticipated regret, this study investigates the factors that affect the use of MFA secure authentication to avoid online account threats. This work extends PMT by showing how the emotion of anticipated regret heightens appraisals of threat and coping. Collapse

Topics: electronic authentication multi-factor authentication IT security mobile system cloud storage

Methods: survey theory development structural equation modeling facial recognition system biometric measurement

Theories: protection motivation theory