AI Decision Making with Dignity? Contrasting Workers’ Justice Perceptions of Human and AI Decision Making in a Human Resource Management Context

2022 | Information Systems Frontiers | Citations: 0

Authors: Bankins, Sarah; Formosa, Paul; Griep, Yannick; Richards, Deborah

Abstract: Using artificial intelligence (AI) to make decisions in human resource managemen ... Expand

Abstract: Using artificial intelligence (AI) to make decisions in human resource management (HRM) raises questions of how fair employees perceive these decisions to be and whether they experience respectful treatment (i.e., interactional justice). In this experimental survey study with open-ended qualitative questions, we examine decision making in six HRM functions and manipulate the decision maker (AI or human) and decision valence (positive or negative) to determine their impact on individuals’ experiences of interactional justice, trust, dehumanization, and perceptions of decision-maker role appropriateness. In terms of decision makers, the use of human decision makers over AIs generally resulted in better perceptions of respectful treatment. In terms of decision valence, people experiencing positive over negative decisions generally resulted in better perceptions of respectful treatment. In instances where these cases conflict, on some indicators people preferred positive AI decisions over negative human decisions. Qualitative responses show how people identify justice concerns with both AI and human decision making. We outline implications for theory, practice, and future research. Collapse

Topics: artificial intelligence decision making individual characteristic missing data job satisfaction

Methods: survey multivariate analysis of variance machine learning computational algorithm decision tree classification

Theories: organizational behavior theory

Development of a machine learning based algorithm to accurately detect schizophrenia based on one-minute EEG recordings

2020 | HICSS | Citations: 0

Authors: Buettner, Ricardo; Beil, David; Scholtz, Stefanie; Djemai, Aadel

Abstract: While diagnosing schizophrenia by physicians based on patients' history and thei ... Expand

Abstract: While diagnosing schizophrenia by physicians based on patients' history and their overall mental health is inaccurate, we report on promising results using a novel, fast and reliable machine learning approach based on electroencephalography (EEG) recordings. We show that a fine granular division of EEG spectra in combination with the Random Forest classifier allows a distinction to be made between paranoid schizophrenic (ICD-10 F20.0) and nonschizophrenic persons with a very good balanced accuracy of 96.77 percent. We evaluate our approach on EEG data from an open neurological and psychiatric repository containing 499 one-minute recordings of n=28 participants (14 paranoid schizophrenic and 14 healthy controls). Since the fact that neither diagnostic tests nor biomarkers are available yet to diagnose paranoid schizophrenia, our approach paves the way to a quick and reliable diagnosis with a high accuracy. Furthermore, interesting insights about the most predictive subbands were gained by analyzing the electroencephalographic spectrum up to 100 Hz. Collapse

Topics: mindfulness personality

Methods: electroencephalography random forest classification machine learning longitudinal research spectral analysis

High-performance detection of epilepsy in seizure-free EEG recordings: A novel machine learning approach using very specific epileptic EEG sub-bands

2019 | International Conference on Information Systems | Citations: 0

Authors: Buettner, Ricardo; Frick, Janek; Rieg, Thilo

Abstract: We applied machine learning to diagnose epilepsy based on the fine-graded spectr ... Expand

Abstract: We applied machine learning to diagnose epilepsy based on the fine-graded spectral analysis of seizure-free (resting state) EEG recordings. Despite using unspecific agglomerated EEG spectra, our fine-graded spectral analysis specifically identified the two EEG resting state sub-bands differentiating healthy people from epileptics (1.5-2 Hz and 11-12.5 Hz). The rigorous evaluation of completely unseen data of 100 EEG recordings (50 belonging to epileptics and the other 50 to healthy people) shows that the approach works successfully, achieving an outstanding accuracy of 99 percent, which significantly outperforms the current benchmark of 70% to 95% by a panel of up to three experienced neurologists. Our epilepsy diagnosis classifier can be implemented in modern EEG analysis devices, especially in intensive care units where early diagnosis and appropriate treatment are decisive in life and death scenarios and where physicians’ error rates are particularly high. Our approach is accurate, robust, fast, and cost-efficient and substantially contributes to Information Systems research in healthcare. The approach is also of high practical and theoretical relevance. Collapse

Methods: electroencephalography computational algorithm random forest classification machine learning longitudinal research

Predicting web site audience demographics using content and design cues

2019 | Information & Management | Citations: 2

Authors: Kim, Iljoo; Pant, Gautam

Abstract: The size and dynamism of the Web pose challenges for all its stakeholders, which ... Expand

Abstract: The size and dynamism of the Web pose challenges for all its stakeholders, which include producers and consumers of content as well as advertisers who want to place advertisements next to relevant content. A critical piece of information for producers/publishers of content as well as advertisers is the demographics of the consumers who are likely to visit a given web site. In this article we explore predictive models that attempt to deduce the demographics of the audience of a web site using cues embedded in the design or the content of its homepage. We find that it is possible to effectively predict different types of demographics of consumers of web sites on the basis of the suggested approach. Through a statistical analysis we observe that several design elements and the content differ significantly among web sites dominated by consumers of different demographic classes. We also suggest the use of an ensemble classifier that combines the content and design cues with the goal of further improving the prediction performance. Collapse

Topics: website web application social network information privacy concern online learning system

Methods: support vector machine logistic regression experiment naïve bayes term frequency–inverse document frequency

Transformational Issues of Big Data and Analytics in Networked Business

2016 | Management Information Systems Quarterly | Citations: 147

Authors: Baesens, Bart; Bapna, Ravi; Marsden, James R.; Vanthienen, Jan; Zhao, J. Leon

Abstract: The era of big data and analytics is upon us and is changing the world dramatica ... Expand

Abstract: The era of big data and analytics is upon us and is changing the world dramatically. The field of Information Systems should be at the forefront of understanding and interpreting the impact of both technologies and management so as to lead the efforts of business research in the big data era. We need to prepare ourselves and our students for this changing world of business. In this discussion, we focus on exploring the technical and managerial issues of business transformation resulting from the insightful adoption and innovative applications of data sciences in business. We end by providing an overview of the papers included in this special issue and outline future research directions. Collapse

Topics: big data marketing management banner advertising decision making advertising management

Methods: field experiment simulation digital trace data experiment qualitative content analysis

Perspective—A New Look at Conflict Management in Work Groups

2016 | Organization Science | Citations: 0

Authors: Carton, Andrew M.; Tewfik, Basima A.

Abstract: Members of work groups are highly interdependent and often share incompatible va ... Expand

Abstract: Members of work groups are highly interdependent and often share incompatible values, objectives, and opinions. As a result, conflict frequently arises. Given the profound impact of conflict on group effectiveness, scholars have sought to identify strategies that can mitigate its downsides and leverage its upsides. Yet research on conflict management strategies has accumulated inconsistent results. In this Perspectives piece, we argue that these inconsistent findings can be resolved if scholars take a more expansive view of the consequences of conflict management strategies: whereas existing research considers how individual strategies influence a single group conflict type (relational, status, process, or task), we consider the impact of individual strategies on all four conflict types. After building a typology by organizing strategies according to the conflict type that each is best equipped to manage, we argue that the strategies most appropriate for managing one type of conflict may systematically backfire by escalating other conflict types. For example, the adoption of a superordinate identity is likely to resolve relational conflict, yet exacerbate status conflict. In addition to uncovering these instances of “negative spillovers,” we shed light on the rarer phenomena of “positive spillovers,” which occur when conflict management strategies resolve conflict types they were not originally designed to influence. By highlighting how individual conflict management strategies influence multiple conflict types—often in contrasting ways—this Perspectives article reconciles conflicting findings and redirects the literature by providing scholars with new recommendations on how to study conflict management in work groups. Collapse

Topics: strategic management accounting

Methods: theory development statistical hypothesis test ethnography decision tree classification triangulation

Theories: conflict theory

Guidelines for Conducting Mixed-methods Research: An Extension and Illustration

2016 | Journal of the Association for Information Systems | Citations: 10

Authors: Venkatesh, Viswanath; Brown, Susan A.; Sullivan, Yulia W.

Abstract: In this paper, we extend the guidelines of Venkatesh et al. (2013) for mixed-met ... Expand

Abstract: In this paper, we extend the guidelines of Venkatesh et al. (2013) for mixed-methods research by identifying and integrating variations in mixed-methods research. By considering 14 properties of mixed-methods research (e.g., purposes, research questions, epistemological assumptions), our guidelines demonstrate how researchers can flexibly identify the existing variations in mixed-methods research and proceed accordingly with a study design that suits their needs. To make the guidelines actionable for various situations and issues that researchers could encounter, we develop a decision tree to map the flow and relationship among the design strategies. We also illustrate one possible type of mixed-methods research in information systems in depth and discuss how to develop and validate metainferences as the outcomes of such a study. Collapse

Topics: technology adoption innovation management information system use knowledge base user resistance

Methods: mixed method survey triangulation decision tree classification qualitative interview

Patterns in Information Systems Portfolio Prioritization: Evidence from Decision Tree Induction

2015 | Management Information Systems Quarterly | Citations: 32

Authors: Karhade, Prasanna; Shaw, Michael J.; Subramanyam, Ramanath

Abstract: Questions pertaining to the locus of information systems (IS) governance have be ... Expand

Abstract: Questions pertaining to the locus of information systems (IS) governance have been extensively examined in existing research. However, questions pertaining to the decision rationale applied for IS portfolio prioritization (why are certain initiatives approved, and why are certain others rejected), noted to be a critical component of IS governance, need further investigation. We submit that the IS strategy of a firm is likely to explain the decision rationale it applies to IS portfolio prioritization and maintain that it is critical to ensure this decision rationale is in congruence with the firm's IS strategy. By extending prior theoretical work on IS strategy types, we develop theoretical profiles of the decision rationale applied to IS portfolio prioritization using three attributes: communicability of decision rationale, consistency in applying decision rationale, and risk appropriateness of decision rationale. Since the decision rationale applied for IS portfolio prioritization is often tacit, unknown even to the decision makers themselves, we employ the decision tree induction methodology to discover this tacit decision rationale. We analyze over 150 IS portfolio prioritization decisions on a multimillion dollar IS portfolio of a multibusiness, Fortune 50 firm and our findings, which support our propositions, indicate that firms that adopt different IS strategies rely on systematically different profiles of decision rationale for IS portfolio prioritization. Implications for IS governance practices are developed. Collapse

Topics: information systems strategy IT governance decision making information technology infrastructure enterprise information system

Methods: decision tree classification computational algorithm field study machine learning theory development

An ambidextrous perspective on business intelligence and analytics support in decision processes: Insights from a multiple case study

2015 | Decision Support Systems | Citations: 21

Authors: Kowalczyk, Martin; Buxmann, Peter

Abstract: Providing data-centric decision support for organizational decision processes is ... Expand

Abstract: Providing data-centric decision support for organizational decision processes is a crucial but challenging task. Business intelligence and analytics (BI&A) equips analytics experts with the technological capabilities to support decision processes with reliable information and analytic insights, thus potentially raising the quality of managerial decision making. However, the very nature of organizational decision processes imposes conflicting task requirements regarding adaptability and rigor. This research proposes ambidexterity as a theoretical lens to investigate data-centric decision support. Based on an in-depth multiple case study of BI&A-supported decision processes, we identify and discuss tensions that arise from the conflicting task requirements and that pose a challenge for effective BI&A support. We also provide insights into tactics for managing these tensions and thus achieving ambidexterity. Additionally, we shed light on the relationship between ambidexterity and decision quality. Integrating the empirical findings from this research, we propose a theory of ambidexterity in decision support, which explains how such ambidexterity can be facilitated and how it affects decision outcomes. Finally, we discuss the study's implications for theory and practice. Collapse

Topics: analytical information system decision making decision support decision quality decision support system

Methods: qualitative interview case study multiple case study survey triangulation

Theories: ambidexterity theory

Web Footprints of Firms: Using Online Isomorphism for Competitor Identification

2015 | Information Systems Research | Citations: 41

Authors: Pant, Gautam; Sheng, Olivia R. L.

Abstract: Competitive isomorphism refers to the phenomenon of competing firms becoming sim ... Expand

Abstract: Competitive isomorphism refers to the phenomenon of competing firms becoming similar as they mimic each other under common market forces. With the growing presence of firms as well as their consumers and suppliers on the Web, we discover a parallel phenomenon of online isomorphism wherein the Web footprints of competing firms are found to overlap. We propose new online metrics based on the content, in-links, and out-links of firms' websites to measure the presence of online isomorphism as well as uncover its utility in predicting competitor relationships. Through rigorous analysis involving more than 2,600 firms, we find that predictive models for competitor identification based on online metrics are largely superior to those using offline data such as Standard Industrial Classification codes and market values of firms. In addition, combining online and offline metrics can boost the predictive performance. We also find that such models are valuable for identifying nuances of competitor relationships such as asymmetry and the role of industrial divisions. Furthermore, the suggested predictive models can effectively rank firms in an industrial division by their likelihood of being competitors to a focal firm as well as identify new future competitors, thus adding to a portfolio of evidence indicating their utility for managers and analysts. Collapse

Topics: website search engine application programming interface web page logistics management

Methods: decision tree classification machine learning logistic regression conceptual modelling experiment