Failures of Fairness in Automation Require a Deeper Understanding of Human-ML Augmentation

MIS Quarterly ◽  
2021 ◽  
Vol 45 (3) ◽  
pp. 1483-1500
Author(s):  
Mike Teodorescu ◽  
◽  
Lily Morse ◽  
Yazeed Awwad ◽  
Gerald Kane ◽  
...  
Keyword(s):  
Machine Learning ◽  
Information Systems ◽  
Current Understanding ◽  
Organizational Processes ◽  
Is Research ◽  
Research Questions ◽  
Human Augmentation

Machine learning (ML) tools reduce the costs of performing repetitive, time-consuming tasks yet run the risk of introducing systematic unfairness into organizational processes. Automated approaches to achieving fair- ness often fail in complex situations, leading some researchers to suggest that human augmentation of ML tools is necessary. However, our current understanding of human–ML augmentation remains limited. In this paper, we argue that the Information Systems (IS) discipline needs a more sophisticated view of and research into human–ML augmentation. We introduce a typology of augmentation for fairness consisting of four quadrants: reactive oversight, proactive oversight, informed reliance, and supervised reliance. We identify significant intersections with previous IS research and distinct managerial approaches to fairness for each quadrant. Several potential research questions emerge from fundamental differences between ML tools trained on data and traditional IS built with code. IS researchers may discover that the differences of ML tools undermine some of the fundamental assumptions upon which classic IS theories and concepts rest. ML may require massive rethinking of significant portions of the corpus of IS research in light of these differences, representing an exciting frontier for research into human–ML augmentation in the years ahead that IS researchers should embrace.

Conduting Feminist Gender Research in the Information Systems Field

2007 ◽  
pp. 1-24 ◽  
Author(s):  
Eileen M. Trauth ◽  
Lynette Kvasny ◽  
Anita Greenhill
Keyword(s):  
Information Systems ◽  
Personal Characteristics ◽  
Feminist Research ◽  
Gender Research ◽  
Is Research ◽  
Research Questions

In this chapter, we explore the methodological and epistemological implications of conducting feminist, gender research in the information systems field. These implications revolve around four core themes: that feminist research is situated in the margins; that current gender and IS research is not adequately problematized; that feminist research questions the legitimacy and appropriateness of positivist research; and that reflection on the personal characteristics of the researcher such as race, gender, sexuality, and class can inform feminist research. We propose four criteria for conducting feminist IS research: (1) engaging in researcher reflexivity; (2) challenging the hegemonic dominance, legitimacy, and appropriateness of positivist epistemologies; (3) theorizing from the margins; and (4) problematizing gender.

scholarly journals Pervasive Lying Posture Tracking

Sensors ◽  
2020 ◽  
Vol 20 (20) ◽  
pp. 5953 ◽  
Author(s):  
Parastoo Alinia ◽  
Ali Samadani ◽  
Mladen Milosevic ◽  
Hassan Ghasemzadeh ◽  
Saman Parvaneh
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Computational Models ◽  
Learning Algorithms ◽  
Pressure Sensors ◽  
Machine Learning Algorithms ◽  
Sensor System ◽  
Accurate Detection ◽  
Research Questions ◽  
Posture Tracking

Automated lying-posture tracking is important in preventing bed-related disorders, such as pressure injuries, sleep apnea, and lower-back pain. Prior research studied in-bed lying posture tracking using sensors of different modalities (e.g., accelerometer and pressure sensors). However, there remain significant gaps in research regarding how to design efficient in-bed lying posture tracking systems. These gaps can be articulated through several research questions, as follows. First, can we design a single-sensor, pervasive, and inexpensive system that can accurately detect lying postures? Second, what computational models are most effective in the accurate detection of lying postures? Finally, what physical configuration of the sensor system is most effective for lying posture tracking? To answer these important research questions, in this article we propose a comprehensive approach for designing a sensor system that uses a single accelerometer along with machine learning algorithms for in-bed lying posture classification. We design two categories of machine learning algorithms based on deep learning and traditional classification with handcrafted features to detect lying postures. We also investigate what wearing sites are the most effective in the accurate detection of lying postures. We extensively evaluate the performance of the proposed algorithms on nine different body locations and four human lying postures using two datasets. Our results show that a system with a single accelerometer can be used with either deep learning or traditional classifiers to accurately detect lying postures. The best models in our approach achieve an F1 score that ranges from 95.2% to 97.8% with a coefficient of variation from 0.03 to 0.05. The results also identify the thighs and chest as the most salient body sites for lying posture tracking. Our findings in this article suggest that, because accelerometers are ubiquitous and inexpensive sensors, they can be a viable source of information for pervasive monitoring of in-bed postures.

scholarly journals The Delphi Method in Information Systems Research (2004-2017)

2019 ◽  
Author(s):  
Ayman Alarabiat ◽  
Isabel Ramos
Keyword(s):  
Information System ◽  
Information Systems ◽  
Delphi Method ◽  
Is Research ◽  
Systems Research ◽  
Information Systems Research ◽  
Innovative Techniques ◽  
Delphi Studies ◽  
Procedural Recommendations ◽  
Do So

After almost 30 years of being used in the information system (IS) discipline, only a few studies have focused on how IS scholars apply the method's guidelines to design Delphi studies. Thus, this paper focuses on the use of the Delphi method in IS research. To do so, articles published between 2004 and 2017 in the Senior IS Scholars' collection of journals of the Association of Information Systems (AIS), describing Delphi studies, were analised. Based on analysis of sixteen (16) retrieved IS studies, we concluded that IS researchers have applied the method’s most important phases and the procedural recommendations to promote rigor were considered in the majority of the analised studies. Nonetheless, IS researchers still need to include detailed information about (1) the steps taken to ensure the validity of the achieved results, (2) better describe the process of selecting and recruiting the experts, and (3) experiment with innovative techniques to keep participants involved in the Delphi process.

scholarly journals Machine Learning and Deep Learning Techniques, Features and Obstacles in the Cataract Diagnosis

2020 ◽  
Vol 9 (3) ◽  
pp. 87-93
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Literature Review ◽  
Congenital Cataract ◽  
Low Cost ◽  
Current Situation ◽  
Learning Techniques ◽  
Research Questions ◽  
New Research

Cataract is a degenerative condition that, according to estimations, will rise globally. Even though there are various proposals about its diagnosis, there are remaining problems to be solved. This paper aims to identify the current situation of the recent investigations on cataract diagnosis using a framework to conduct the literature review with the intention of answering the following research questions: RQ1) Which are the existing methods for cataract diagnosis? RQ2) Which are the features considered for the diagnosis of cataracts? RQ3) Which is the existing classification when diagnosing cataracts? RQ4) And Which obstacles arise when diagnosing cataracts? Additionally, a cross-analysis of the results was made. The results showed that new research is required in: (1) the classification of “congenital cataract” and, (2) portable solutions, which are necessary to make cataract diagnoses easily and at a low cost.

scholarly journals Construction of design science research questions

2020 ◽  
Author(s):  
NH Thuan ◽  
Andreas Drechsler ◽  
Pedro Antunes
Keyword(s):  
Information Systems ◽  
Design Science ◽  
Science Research ◽  
Design Science Research ◽  
Fundamental Step ◽  
Research Questions

© 2019 by the Association for Information Systems. Posing research questions represents a fundamental step to guide and direct how researchers develop knowledge in research. In design science research (DSR), researchers need to pose research questions to define the scope and the modes of inquiry, characterize the artifacts, and communicate the contributions. Despite the importance of research questions, research provides few guidelines on how to construct suitable DSR research questions. We fill this gap by exploring ways of constructing DSR research questions and analyzing the research questions in a sample of 104 DSR publications. We found that about two-thirds of the analyzed DSR publications actually used research questions to link their problem statements to research approaches and that most questions focused on solving problems. Based on our analysis, we derive a typology of DSR question formulation to provide guidelines and patterns that help researchers formulate research questions when conducting their DSR projects.

Process architecture development using Riva and ARIS: comparative study

2018 ◽  
Vol 24 (3) ◽  
pp. 837-858 ◽  
Author(s):  
Dina Tbaishat
Keyword(s):  
Information Systems ◽  
Comparative Study ◽  
Academic Libraries ◽  
Theory And Practice ◽  
Organizational Processes ◽  
Content Type ◽  
Big Picture ◽  
Generic Library ◽  
The Right ◽  
Process Architecture

Purpose Business process modeling has been given great attention due to its crucial role in developing computer-based systems that support (and automate) organizational processes. In information systems, building the right process architecture is vital, since a poor division of organizational processes can lead to complex designs or incoherent structure. Moreover, process architecture acts as a “big picture” of what the organization does, and represents dynamic relationships between the existing processes, which, in turn, helps understand how the organization works (Ould, 2005). The paper aims to discuss these issues. Design/methodology/approach This paper discusses the derivation of a process architecture diagram (PAD) using the Riva method in detail, in contrast to the PAD developed using Architecture of Integrated Information Systems (ARIS). The information system selected as an example for this comparative study is in the context of academic libraries, embedding various yet generic library processes. Findings Organizational process architecture provides a basis for business management, based on specific framework supported by integrated tools and methods (Kozina, 2006). ARIS and Riva process architecture methods can be used to visualize and document organizational processes. Understanding the merits and weaknesses of each method is essential to identify possible areas of application. Research limitations/implications The processes selected were generic in most academic libraries not taking into consideration special personalized processes. These can be added though. Practical implications Both techniques are feasible and can be used to view and analyze library processes, allowing bridging the gap between theory and practice. Originality/value A number of process architecture methods are available, however, few studies focused on assessing these methods, and comparing some of them to find out how easy they are to be used in particular contexts, and whether they can be standardized.

Social Drivers for ICT Diffusion among Agrarian Business Enterprises in Nigeria

2013 ◽  
pp. 91-104
Author(s):  
Bartholomew Aleke ◽  
Udechukwu Ojiako ◽  
David Wainwright
Keyword(s):  
Developing Countries ◽  
Social Network ◽  
Information Systems ◽  
Diffusion Of Innovation ◽  
Social Network Theory ◽  
Conceptual Basis ◽  
Diffusion Of Innovation Theory ◽  
Is Research ◽  
Business Enterprises ◽  
Successful Innovation

In this paper, the authors examine the process of diffusion of innovation ICT within agrarian business enterprises operating in developing countries. There is substantial research in the area of Diffusion of Innovation Theory (DoI) and its application to Information Systems (IS) research within organisations. However, in recognition of the conceptual limitations of DoI, researchers have called for the incorporation of aspects of Social Network Theory (SNT) into DoI frameworks. The findings of this research suggest that an understanding of the conceptual basis of innovation is a major driver of successful innovation adoption.

Information Systems Evaluation

2013 ◽  
pp. 333-354 ◽  
Author(s):  
Si Chen ◽  
Nor Mardziah Osman ◽  
Guo Chao Alex Peng
Keyword(s):  
Information Systems ◽  
Formative Evaluation ◽  
Current System ◽  
Evaluation Research ◽  
Evaluation Methodologies ◽  
Is Implementation ◽  
Is Research ◽  
Is Evaluation ◽  
Evaluation Approaches ◽  
Goals And Objectives

Due to the prevalent use of Information Systems (IS) in modern organisations, evaluation research in this field is becoming more and more important. In light of this, a set of rigorous methodologies were developed and used by IS researchers and practitioners to evaluate the increasingly complex IS implementation used. Moreover, different types of IS and different focusing perspectives of the evaluation require the selection and use of different evaluation approaches and methodologies. This chapter aims to identify, explore, investigate, and discuss the various key methodologies that can be used in IS evaluation from different perspectives, namely in nature (e.g. summative vs. formative evaluation) and in strategy (e.g. goal-based, goal-free, and criteria-based evaluation). Six case studies are also presented and discussed in this chapter to illustrate how the different IS evaluation methodologies can be applied in practices. The chapter concludes that evaluation methodologies should be selected depending on the nature of the IS and the specific goals and objectives of the evaluation. Nonetheless, it is also proposed that formative criteria-based evaluation and summative criteria-based evaluation are currently among the more widely used in IS research. The authors suggest that the combined used of one or more of these approaches can be applied at different stages of the IS life cycle in order to generate more rigorous and reliable evaluation outcomes. Moreover, results and outcomes of IS evaluation research will not just be useful in practically guiding actions to improve the current system, but can also be used to generate new knowledge and theory to be adopted by future IS research.

Sign in / Sign up

Export Citation Format

Share Document