67471 research outputs found
Sort by
The Algorithm and the Crowd: A Design Science Research Approach to the Democratization of Algorithmic Management
Get PDFAlgorithmic Management (AM) is often criticized for its opacity and centralized decision-making power, thereby limiting workers’ autonomy and participation. Building on emerging work in participatory AM and drawing on established literature on democratization mechanisms, we adopt a Design Science Research approach to develop a prototype democratized AM system for workforce scheduling. Our artifact works to ensure that workers’ voices are meaningfully represented in algorithmic decision-making by building upon three democratization mechanisms: preference, feedback, and transparency. The resulting prototype is designed based on requirements derived from those democratization mechanisms and receives high worker satisfaction in a qualitative evaluation. We extend existing research by embedding democratization principles into a functioning AM system and derive design requirements that can inform the development of future AM systems. In doing so, we demonstrate how democratization can be effectively embedded in algorithmic systems in ways that are both effective and valued by workers
Generating Value in Data Ecosystems: The Ontological Characteristics of Data as Enablers of Data-driven Value Propositions
Get PDFThe unique characteristics of data require rethinking how firms develop new value propositions. Viewing data as a static resource, as current information systems (IS) literature does, is insufficient; instead, it must be seen as a dynamic, co-constructed asset embedded in evolving data ecosystems. This study explores how data characteristics enable such value propositions through a qualitative case study of a healthcare infrastructure provider. Findings reveal a dual model: a platform supporting care delivery, and a data-driven proposition offering insights from patient-generated data. Thematic analysis identifies four core data characteristics (recontextualization, editability, portability, and non-rivalry) each linked to empirical patterns that explain how data underpins value generation. The study contributes to the literature on data ecosystems by revealing the ontological characteristics of data as enablers of data-driven value propositions, while offering practical guidance for firms leveraging digital data for strategic advantage
The Role of Alcohol Availability and Community-level Factors on Fatal and Non-Fatal Overdoses
Get PDFThis study explores the dynamic interplay between alcohol and community health outcomes, focusing on the influence of alcohol outlets on fatal and non-fatal overdose risk across urban spaces. Central to this discourse is the recognition of the pivotal impact of opioids on drug overdose incidents within communities. Existing literature has laid the groundwork for understanding the complex web of factors contributing to overdose fatalities. However, gaps persist in our understanding of how the distribution of alcohol outlets across urban spaces interacts with other contextual variables to shape overdose risk at the local level. By employing geospatial techniques, this research seeks to bridge these gaps, shedding light on substance-related harms and informing targeted interventions tailored to specific communities in Milwaukee, Wisconsin
Mapping the Moral Foundations of Machines: A Vignette-Based Inquiry into Moral Reasoning Across Six Large Language Model Platforms
Get PDFLarge language models take on an ever-increasing role in our societies. In particular, they can guide and inform human decisions. Understanding the moral profiles of LLMs, whether they are stable within the same models over time and across models, is key to ensuring LLM-informed decisions are not unduly shaped by model-specific moral emphases. Drawing from Moral Foundations Theory (MFT), we evaluated the moral profiles of six different LLMs developed in different sociotechnical contexts to measure stability of moral profiles across models. We measured test-retest reliability of LLMs’ explanations of their moral judgments using content analysis to test the stability of output patterns within LLMs. We found mostly stable moral profiles within and across models with few exceptions. We frame these cross-model differences and exceptions as actionable guidance for model selection and routing in morally relevant use cases
Improving LLM Interpretability with User-Centric Chain-of-Thought Reasoning
Get PDFAdvancing reasoning capabilities allow large language models (LLMs) to tackle increasingly complex problems, while reasoning traces—intermediate steps toward solutions—open up high-stakes applications by enabling human inspection of AI decision-making. However, current approaches prioritize model performance over human interpretability, limiting effective human-AI collaboration. In this study, we design and evaluate a human-centered approach that structures reasoning traces based on self-contained, verifiable steps, enabling users to independently assess and correct AI reasoning. Our approach uses XML-like tags to encode reasoning content and metadata, facilitating targeted feedback. Evaluation on mathematical reasoning tasks shows our approach maintains equivalent performance to standard Chain-of-Thought reasoning while enhancing interpretability. User studies demonstrate significant improvements in perceived usefulness and ease of use. This work advances understanding of how user-centric design of LLM outputs can better serve human collaboration needs in high-stakes AI deployments
Edge-Kalman Filter for Improved Quality of Service in Indoor Spatial Temporal Systems
Get PDFIndoor Spatial Temporal Systems (ISTS) enable real-time location-aware services in applications like indoor navigation, healthcare monitoring, and emergency response. They rely on technologies such as Bluetooth, Wi-Fi, and Zigbee to estimate distances from Received Signal Strength Indicator (RSSI) signals, which are prone to aleatoric uncertainty. Traditional ISTS mitigate this by applying noise filtering at the cloud layer but introduces transmission latency. Recent work shifts filtering to the edge to reduce transmission latency. However, frequent noise parameter updates in the filtering process lead to a high computational load, which increases communication latency and degrades the Quality of Service (QoS). To address this, we propose a lightweight edge-Kalman Filter (eKF) that updates noise parameters only when a significant change in density ratio is detected using the Kullback-Leibler Importance Estimation Procedure (KLIEP). Experimental results show that eKF reduces computational overhead, enhances distance estimation accuracy, and scales across technologies, demonstrating its potential to improve real-time QoS in ISTS
Beyond Accessibility: How Intelligent Assistive Technologies Improve Activities of Daily Life for Visually Impaired People in South Africa
Get PDFOur study explores how intelligent assistive technologies (IATs) can enable visually impaired people (VIPs) to overcome barriers to inclusion in a digital society to ultimately improve their quality of life. Drawing on the Social Model of Disability (SMD), which frames disability as a consequence of social and institutional barriers rather than individual impairments, we employ semi-structured interviews and an online qualitative survey with n=61 VIPs in South Africa. Using descriptive statistics and Qualitative Comparative Analysis (QCA), we uncover nine configurations, clustered along three broader combinations of conditions, that support and hinder IAT-mediated inclusion. Most notably, we identify that autonomy of VIPs and accessibility of IATs are primary predictors of IAT’s ability to achieve social participation. Our findings contribute to Information Systems (IS) literature at the intersection of technology and social participation. We further formulate implications for research and policymakers to foster social inclusion of VIPs in the Global South
Detection of Contradictions and Inconsistencies in Regulatory Documents using Prompt-Engineering
Get PDFCompanies create regulatory documents, such as policies, standards, and guidelines, to define their processes and structures. Frequent updates to these documents can lead to inconsistencies and contradictions between the respective regulations, which can result in errors, delays, asset compromise, or enabling fraud and non-compliance. Given the variety of document types and their thematic, structural, lexical, syntactic, and domain-specific differences, automated conflict detection remains a challenge, especially due to the lack of annotated data from practice. As an alternative to supervised approaches, this paper investigates whether a prompt-based classifier can detect contradictions and inconsistencies between regulatory texts and what level of accuracy can be achieved. The evaluation of three prompt variants and seven large language models on a real-world regulatory dataset shows that the detection accuracy of a prompt-based classifier (F1- score of 0.851), which includes 26 detailed formulated rules, is only 1.29% lower than that of a supervised model (F1-score of 0.862) trained with annotated data
Undercollateralized Lending with Inverum DeFi Protocol
Get PDFDecentralized Finance (DeFi) enables financial services to operate without centralized intermediaries, using smart contracts and blockchain consensus to ensure transparency and trust minimization. While DeFi protocols like Aave and MakerDAO use overcollateralization to mitigate credit risk, this approach creates capital inefficiencies and limits access to borrowers lacking on-chain assets. This paper introduces Inverum, a novel DeFi lending protocol designed to support undercollateralized loans for Web3 businesses and Decentralized Autonomous Organizations (DAOs). Inverum integrates on-chain credit scoring via soulbound tokens, decentralized liquidity pools, and governance-driven incentives to enable trustless, reputation-based lending. The protocol offers a fully composable framework for exploring undercollateralized lending without relying on traditional identity or off-chain reputation systems, contributing a research-ready model for future experimentation and protocol design