A Multi-Label Classification with an Adversarial-Based Denoising Autoencoder for Medical Image Annotation

2023 | ACM Transactions on Management Information Systems | Citations: 0

Authors: Chai, Yidong; Liu, Hongyan; Xu, Jie; Samtani, Sagar; Jiang, Yuanchun; Liu, Haoxin

Abstract: Medical image annotation aims to automatically describe the content of medical i ... Expand

Abstract: Medical image annotation aims to automatically describe the content of medical images. It helps doctors to understand the content of medical images and make better informed decisions like diagnoses. Existing methods mainly follow the approach for natural images and fail to emphasize the object abnormalities, which is the essence of medical images annotation. In light of this, we propose to transform the medical image annotation to a multi-label classification problem, where object abnormalities are focused directly. However, extant multi-label classification studies rely on arduous feature engineering, or do not solve label correlation issues well in medical images. To solve these problems, we propose a novel deep learning model where a frequent pattern mining component and an adversarial-based denoising autoencoder component are introduced. Extensive experiments are conducted on a real retinal image dataset to evaluate the performance of the proposed model. Results indicate that the proposed model significantly outperforms image captioning baselines and multi-label classification baselines. Collapse

Topics: workstation

Methods: data transformation autoencoder experiment deep learning literature study

Token Incentives in a Volatile Crypto Market: The Effects of Token Price Volatility on User Contribution

2023 | Journal of Management Information Systems | Citations: 0

Authors: Chen, Kun; Fan, Yifan; Liao, Shaoyi Stephen

Abstract: Crypto tokens, issued and managed via smart contracts, function as rewards in bl ... Expand

Abstract: Crypto tokens, issued and managed via smart contracts, function as rewards in blockchain systems to encourage user participation. Distinct from monetary incentives, token incentives are uncertain in reward magnitude due to the large swings in token prices on crypto markets. By focusing on token price volatility, this study investigates how the reward uncertainty affects user contribution in a tokenized digital platform. Our empirical setting is Steemit, a platform where bloggers write posts and share token rewards based on their posts' popularity. We find that while high token price volatility induces a large volume of blog posts, it diminishes post quality. The dichotomous effects are explained via two mechanisms: users' direct reactions to reward uncertainties and their indirect reactions, mediated by altered token preference amid volatility shocks. Deeply exploring this dynamic process, our results reveal that token price volatility facilitates a platform's network short-term effect but impairs long-term user creativity. Our empirical findings thus extend the literature on blockchain economics and cryptocurrencies and have practical implications for the design of incentive mechanisms on tokenized digital platforms. Collapse

Topics: blog blockchain digital platform cryptocurrency user behavior

Methods: longitudinal research instrumental variables estimation data transformation principal component analysis vector autoregression

Theories: motivation crowding theory

Navigating AI in Personnel Selection: A Scenario-based Study on Applicants' Perceptions

2023 | International Conference on Information Systems | Citations: 0

Authors: Czernietzki, Charlotte; Märtins, Julian; Westmattelmann, Daniel; Grotenhermen, Jan-Gerrit; Oldeweme, Andreas; Baumeister, Viktoria; Schewe, Gerhard

Abstract: AI-based systems are increasingly deployed on organizational tasks, such as per ... Expand

Abstract: AI-based systems are increasingly deployed on organizational tasks, such as personnel selection decisions. As existing research indicates that applicants generally react negatively to the use of AI in personnel selection, this study examines how organizations can mitigate adverse reactions to fully exploit the benefits of AI. To obtain robust results, we recruited an online sample of German participants (N = 1,852) and presented them with various selection scenarios. Using a between-subject design, the process stage (preselection vs. interview) and the degree of process automation (augmented vs. automated) were manipulated. By employing a multidimensional conceptualization of transparency, we show that disclosure and accuracy positively impact procedural justice perceptions, a strong predictor of process quality assessment. This relationship is robust across selection contexts. Results indicate that applicants prefer AI for pre-selection and as human decision support, thus offering overall insights into design choices for AI in selection, optimizing applicant reactions. Collapse

Topics: artificial intelligence business process automation data quality electronic mail virtual reality

Methods: qualitative interview partial least squares path modeling survey parametric test ANOVA

Theories: procedural justice theory organizational justice theory

Corporate Users’ Attachment to Social Networking Sites: Examining the Role of Social Capital and Perceived Benefits

2023 | Information Systems Frontiers | Citations: 0

Authors: Dhar, Suparna; Bose, Indranil

Abstract: The debate on the pros and cons of employee attachment to social networking site ... Expand

Abstract: The debate on the pros and cons of employee attachment to social networking sites (SNS) has led to social media policy paralysis in many organizations, and often a prohibition on employee use of SNS. This paper examines corporate users’ attachment to SNS. An analysis of 316 survey responses showed that corporate users’ socialization in large public SNS was steeped in perceived work-related benefits, which in turn nourished their SNS attachment. Social use outperformed informational use in generating perceived work-related benefits from SNS. Weak ties in large heterogeneous networks resulted in strategic and operational benefits, whereas the effects of strong bonding in homogenous networks were limited to operational benefits. The paper contributes to research on SNS use by corporate users and the debate on the effect of SNS use for work. The findings will benefit SNS strategists of organizations and policymakers to exploit the benefit potential of public SNS. Collapse

Topics: social capital Facebook LinkedIn information system use accounting

Methods: survey data transformation principal component analysis path analysis structural equation modeling

Theories: uses and gratifications theory social capital theory attachment theory

Variations in the Corporate Social Responsibility-Performance Relationship in Emerging Market Firms

2023 | Organization Science | Citations: 0

Authors: Cuervo-Cazurra, Alvaro; Purkayastha, Saptarshi; Ramaswamy, Kannan

Abstract: Corporate social responsibility (CSR) and its impact on performance have generat ... Expand

Abstract: Corporate social responsibility (CSR) and its impact on performance have generated a debate that has evolved across several perspectives (shareholder, stakeholder, resource-based, and contingency). Building on the resource-based and contingency perspectives, we shed new light on this debate by analyzing the impact of CSR on performance in emerging market firms, advancing the idea that CSR is a mechanism that helps address market and government failures. We first argue that CSR’s three constituent dimensions (environmental, social, and governance) vary in their impact on performance because each dimension has a different mitigating effect on contextual failures that hobble emerging market firms. Specifically, we contend that social CSR has a larger effect on performance than either governance CSR or environmental CSR for emerging market firms, because the former helps build capabilities that more directly reduce the negative consequences of government failures in the provision of public goods and services that firms need to operate efficiently. We then provide additional depth to this idea by arguing that other mechanisms used for mitigating market failures in an emerging market context, namely firm-level business group affiliation and country-level government policy nudges, strengthen this differential influence of each of the three dimensions of CSR on performance. Analyses of a sample of 89 publicly traded Indian firms from 2007 to 2017 support these arguments. Funding: A. Cuervo-Cazurra thanks the Lloyd Mullin fellowship at Northeastern University for financial support. S. Purkayastha thanks Indian Institute of Management, Calcutta for financial support. Supplemental Material: The online appendix is available at https://doi.org/10.1287/orsc.2022.1639. Collapse

Topics: organizational productivity database system public good legal context societal evaluation criteria

Methods: experimental group longitudinal research robustness check instrumental variables estimation difference in differences

Theories: resource based view of the firm

A Task-Technology-Identity Fit Model of Smartwatch Utilisation and User Satisfaction: A Hybrid SEM-Neural Network Approach

2023 | Information Systems Frontiers | Citations: 0

Authors: El-Masri, Mazen; Al-Yafi, Karim; Kamal, Muhammad Mustafa

Abstract: Smartwatches are wearable devices intended to be smartphone companions that capt ... Expand

Abstract: Smartwatches are wearable devices intended to be smartphone companions that capture health data and ease access to notifications. They have also become personalisable standing as a fashion statement. This combination resulted in staggering adoption rates recently leading to question whether smartwatch users’ choice and use satisfaction emerge from utility features or from its fashion characteristics. This paper proposes and validates a fit theory to investigate the antecedents of adopters’ satisfaction. Besides evaluating fit with identity, the model assesses both perceived and actual task-technology fit of smartwatches. A questionnaire-based quantitative approach is used to collect data from about 300 smartwatch users in Qatar. To test the proposed model, data is analysed using structural equation modeling (SEM) and artificial neural networks (ANN). Furthermore, ANN sensitivity analysis ranks the importance of the fit factors affecting users’ choice during pre- and post-adoption stages. Both task-technology and technology-identity fit factors are quasi-equally important in explaining 62% of satisfaction variance. ANN analysis revealed that post-adoption satisfaction is primarily attributed to smartwatches’ ability to fit with users’ identity and secondarily to its perceived fit with tasks. Nevertheless, pre-adoption choice of smartwatches is mainly guided by their functionality. This paper is the first to propose and validate an integrated task-technology-identity fit model to explain smartwatch utilization and users’ satisfaction. The originality also lies in assessing actual task-technology fit and as perceived by users. Employing two modes of analysis revealed extra insights too. Collapse

Topics: smart watch mobile system user satisfaction personality smart wearable device

Methods: structural equation modeling data transformation survey factor analysis principal component analysis

Theories: theory of task technology fit unified theory of acceptance and use of technology technology acceptance model

Is Academic Integrity at Risk? Perceived Ethics and Technology Acceptance of ChatGPT

2023 | Americas Conference on Information Systems | Citations: 0

Authors: Farrell, Wendy; Bogodistov, Yevgen; Mössenlechner, Claudia

Abstract: The academic community is experiencing a significant disruption thanks to a new ... Expand

Abstract: The academic community is experiencing a significant disruption thanks to a new technology called ChatGPT. There is news almost daily about how Large Language Models technologies have the potential to allow students to cheat undetected. This leads many universities to discuss ways to regulate and even ban its use. But is this necessary? The previous research focuses mainly on universities' and lecturers' views but fails to consider the students' views. Do students even want to use the system, and if they do, will they use it for nefarious purposes? Extending the Technology Acceptance Model with Perceived Ethics, this quantitative study researched how students perceive ChatGPT from an ethical perspective and how that impacted their intention to use the system for school work. 277 Students from 40 countries took part in the survey. Results indicated that Perceived Ethics has a direct positive influence on the attitude toward usage but only an indirect effect on the behavioral intention to use the system for school work. This study is one of the first to consider how students perceive ChatGPT from an ethical perspective and how that impacts their perception of and intention to use the Large Language Models technology system. Collapse

Topics: usage intention perceived usefulness behavioral intention perceived ease of use information system use

Methods: chatGPT data transformation survey PLS tool partial least squares regression

Theories: technology acceptance model unified theory of acceptance and use of technology

Near-Infrared Spectroscopy for Bladder Monitoring: A Machine Learning Approach

2023 | ACM Transactions on Management Information Systems | Citations: 0

Authors: Fechner, Pascal; König, Fabian; Kratsch, Wolfgang; Lockl, Jannik; Röglinger, Maximilian

Abstract: Patients living with neurogenic bladder dysfunction can lose the sensation of th ... Expand

Abstract: Patients living with neurogenic bladder dysfunction can lose the sensation of their bladder filling. To avoid over-distension of the urinary bladder and prevent long-term damage to the urinary tract, the gold standard treatment is clean intermittent catheterization at predefined time intervals. However, the emptying schedule does not consider actual bladder volume, meaning that catheterization is performed more often than necessary, which can lead to complications such as urinary tract infections. Time-consuming catheterization also interferes with patients' daily routines and, in the case of an empty bladder, uses human and material resources unnecessarily. To enable individually tailored and volume-responsive bladder management, we design a model for the continuous monitoring of bladder volume. During our design science research process, we evaluate the model's applicability and usefulness through interviews with affected patients, prototyping, and application to a real-world in vivo dataset. The developed prototype predicts bladder volume based on relevant sensor data (i.e., near-infrared spectroscopy and acceleration) and the time elapsed since the previous micturition. Our comparison of several supervised state-of-the-art machine and deep learning models reveals that a long short-term memory network architecture achieves a mean absolute error of 116.7 ml that can improve bladder management for patients. Collapse

Topics: computer hardware open source mobile application user interface programming language

Methods: design artifact qualitative interview design process support vector machine computational algorithm

Law, Economics, and Privacy: Implications of Government Policies on Website and Third-Party Information Sharing

2023 | Information Systems Research | Citations: 0

Authors: Gopal, Ram D.; Hidaji, Hooman; Kutlu, Sule Nur; Patterson, Raymond A.; Yaraghi, Niam

Abstract: Widespread abuse of internet users’ privacy online has prompted user advocacy gr ... Expand

Abstract: Widespread abuse of internet users’ privacy online has prompted user advocacy groups to implore governments to intervene and protect consumer rights. To study such interventions’ effects, we examine data-protection policies that policy makers and governments can enforce on websites, including consent-based user information sharing and subsidizing competing websites. We use a stylized analytical model to examine such policies’ impact on the decisions and outcomes of websites, users, and third-parties. Interestingly, we find that even though a consent-based policy may improve user surplus, in the absence of market entry and exit (static market), it has the unintended consequence of increasing the number of third-parties and, thus, sharing of user information. We also determine that both consent-based and website subsidization policies may reduce competition by driving websites out of the market—to the detriment of user surplus and social welfare. To validate our analytical model’s findings, we empirically investigated the impact of a consent-based policy on third-parties in a natural experiment of the California Consumer Privacy Act. These findings raise significant implications for policy making surrounding online privacy. History: Giri Kumar Tayi, Senior Editor; Nachiketa Sahoo, Associate Editor. Funding: Financial support from the Social Sciences and Humanities Research Council of Canada is gratefully acknowledged. R. D. Gopal gratefully acknowledges support by the Gillmore Centre for Financial Technology at the Warwick Business School. Supplemental Material: The online appendix is available at https://doi.org/10.1287/isre.2022.1178. Collapse

Topics: website information privacy concern data protection privacy legal context

Methods: mathematical model natural experiment experimental group longitudinal research time series analysis

Co-evolution of neural architectures and features for stock market forecasting: A multi-objective decision perspective

2023 | Decision Support Systems | Citations: 0

Authors: Hafiz, Faizal; Broekaert, Jan; La Torre, Davide; Swain, Akshya

Abstract: In a multi-objective setting, a portfolio manager’s highly consequential decisio ... Expand

Abstract: In a multi-objective setting, a portfolio manager’s highly consequential decisions can benefit from assessing alternative forecasting models of stock index movement. The present investigation proposes a new approach to identify a set of non-dominated neural network models for further selection by the decision-maker. A new coevolution approach is proposed to simultaneously select the features and topology of neural networks (collectively referred to as neural architecture), where the features are viewed from a topological perspective as input neurons. Further, the co-evolution is posed as a multi-criteria problem to evolve sparse and efficacious neural architectures. The well-known dominance and decomposition based multi-objective evolutionary algorithms are augmented with a non-geometric crossover operator to diversify and balance the search for neural architectures across conflicting criteria. Moreover, the co-evolution is augmented to accommodate the data-based implications of distinct market behaviors prior to and during the ongoing COVID-19 pandemic. A detailed comparative evaluation is carried out with the conventional sequential approach of feature selection followed by neural topology design, as well as a scalarized co-evolution approach. The results on three market indices (NASDAQ, NYSE, and S&P500) in pre- and peri-COVID time windows convincingly demonstrate that the proposed co-evolution approach can evolve a set of non-dominated neural forecasting models with better generalization capabilities. Collapse

Topics: covid 19 pandemic decision support system decision making

Methods: computational algorithm artificial neural network data transformation machine learning principal component analysis

Theories: co-evolution theory