Search

Major shocks such as the COVID-19 pandemic create unique and exceptional challenges for different entities, including individuals, groups, and organizations. In this special issue editorial, we introduce the concept of digital resilience, which refers to the capabilities developed through the use of digital technologies to absorb major shocks, adapt to disruptions caused by the shocks, and transform to a new stable state, where entities are more prepared to deal with major shocks. The individual papers in this special issue offer compelling examples of how digital resilience is exhibited and how the process of digital resilience can unfold in response to specific major shocks. Drawing upon and extending these papers, we present an integrated framework of how digital technology can help build resilience capabilities, which is missing in past research but needed to mitigate and manage future major shocks, including financial recessions and climate change. We conclude with four important themes for future IS research. [ABSTRACT FROM AUTHOR]

Modern artificial intelligence (AI) is heavily reliant on deep learning (DL), an emerging class of algorithms that can automatically detect non-trivial patterns from petabytes of rapidly evolving "Big Data." Although the information systems (IS) discipline has embraced DL, questions remain about DL's interface with a domain and theory and DL contribution types. In this paper, we present a DL information systems research (DL-ISR) schematic that reviews DL while considering the role of the application environment and knowledge base, summarizes extant DL research in IS, a knowledge contribution framework (KCF) to position DL contributions, and ten guidelines to help IS scholars design, execute, and present DL for computational, behavioral, or economic IS research. We illustrate a research contribution to DL for cybersecurity. This article's contribution to theory resides in the conceptual DL-ISR schematic and KCF, while its contributions to practice are based on its practical guidelines for executing DL-based projects. [ABSTRACT FROM AUTHOR]

Digital platforms have replaced traditional markets in most industries and orchestrate socioeconomic aspects of our lives. We address the problem of negative direct side network effects that arise with an increased number of agents on one side of the platform. Negative effects, if unaddressed, lead to undesired long-term consequences for the platform by developing a positive vicious cycle. Addressing negative effects require dynamic solution mechanisms that adapt to the changing landscape of platforms. The recommender systems literature has proposed multi-sided recommender systems (MSR) as a dynamic solution to many problems on platforms. However, current state-of-the-art MSRs do not consider uncertainty in predicting agents' choices, resulting in limited efficacy. We present a robust multi-sided recommender system that considers estimation errors in agents' choice to address this concern. Extensive experiments with agent-based models—ride-pooling and education platform—provide support for the efficacy and generalizability of the robust MSR to address negative effects. [ABSTRACT FROM AUTHOR]

This paper presents methods to proactively choose individuals to test for infection during a pandemic such as COVID-19, characterized by high contagion and presence of asymptomatic carriers. We show that by a smart integration of exploration/exploitation balancing, contact tracing, and location-based sampling, one can effectively mitigate the disease spread and significantly reduce the infection rates and death rates. Under different vaccination policies and under different compliance levels to quarantine order and/or testing requests, our smart testing algorithm can bring down the death rate significantly by 20% to 30%, as well as the percentage of infected drops by approximately 30%. The load on hospitals at peak times, a crucial aspect of managing COVID-19, drops, by 50% when implementing smart testing. We also show how procedural fairness can be incorporated into our method and present results that show that this can be done without hurting the effectiveness of the mitigation that can be achieved. This paper presents methods to choose individuals to test for infection during a pandemic such as COVID-19, characterized by high contagion and presence of asymptomatic carriers. The smart-testing ideas presented here are motivated by active learning and multi-armed bandit techniques in machine learning. Our active sampling method works in conjunction with quarantine policies, can handle different objectives, and is dynamic and adaptive in the sense that it continually adapts to changes in real-time data. The bandit algorithm uses contact tracing, location-based sampling and random sampling in order to select specific individuals to test. Using a data-driven agent-based model simulating New York City we show that the algorithm samples individuals to test in a manner that rapidly traces infected individuals. Experiments also suggest that smart-testing can significantly reduce the death rates as compared with current methods, with or without vaccination. While smart testing strategies can help mitigate disease spread, there could be unintended consequences with fairness or bias when deployed in real-world settings. To this end we show how procedural fairness can be incorporated into our method and present results that show that this can be done without hurting the effectiveness of the mitigation that can be achieved. History: Ahmed Abbasi, Senior Editor; Maytal Saar-Tsechansky, Associate Editor. Funding: W. Yahav is supported by the Jeremy Coller Foundation and the Henry Crown Institute of Business Research in Israel. Supplemental Material: The e-companion is available at https://doi.org/10.1287/isre.2023.1215. [ABSTRACT FROM AUTHOR]

Abstract. Fraud, waste, and abuse are significant problems in major industries such as healthcare, particularly when third-party payers such as Medicare are involved. Auditors looking for fraudulent activities use scoring [...]

An introduction is presented in which the editors discuss various topics including a three-staged process for examining the papers (articles) that author submit to the journal, the role of machine learning in information systems research, and artificial intelligence.

An introduction is presented in which the guest editors discuss the theme of the issue, a pluralistic approach in advancing research transparency at MIS Quarterly Journal.

This paper presents and synthesizes results from three studies (two controlled experiments and one interview) on using recommender systems to reduce healthcare costs at prescription time, while taking time pressure into account. All of our subjects were real practicing physicians, nurse practitioners, or physician assistants. Across these studies, a total of 160 medical practitioners used a system that provides recommendations for medications along with associated cost information. The main finding was a general tendency among practitioners to reduce healthcare costs by prescribing lower cost medications when cost information is provided by a recommender system. The time pressure faced daily by prescribers, however, appears to impact the use of recommendations by nurse practitioners and physician assistants more than it does physicians. These results have significant implications for cost reduction in healthcare and for the design of effective real-time healthcare recommender systems. [ABSTRACT FROM AUTHOR]

We study digital multisided platforms as complex adaptive business systems (CABS) where multiple sides have different and evolving objectives, preferences, and constraints. CABS are characterized by irreducible uncertainty, which cannot be reduced by the traditional approaches of collecting and processing data. Irreducible uncertainty in the system gives rise to a complex search matching problem between agents and value enhancing transactions. This paper presents a recommender systems-based approach for taming the complexity by allowing agents to coevolve and learn in the system. We propose a novel two-sided recommender system framework, which considers emergence on both sides of the platform and adapts to the changing environment to influence agents. An agent-based simulation model is developed based on popular internet-based educational platforms to study this complex system and test our hypotheses. Our results show the value of a twosided recommender system to tame complex search matching in platforms. We discuss implications for information systems and complexity science research. [ABSTRACT FROM AUTHOR]

We address the discovery of periodic patterns in sequence data. Building on prior work in this area, we present definitions and new methods for characterizing and identifying four types of [...]

Using an ensemble of models often results in better performance than using a single 'best' model. We present a new approach based on data envelopment analysis (DEA) for model combination. […]

A drawback of traditional data-mining methods is that they do not leverage prior knowledge of users. In prior work, we proposed a method that could discover unexpected patterns in data by using domain knowledge in a systematic manner. In this paper, we present new methods for discovering a minimal set of unexpected patterns by combining the two independent concepts of minimality and unexpectedness, both of which have been well-studied in the KDD literature. We demonstrate the strengths of this approach experimentally using a case study in a marketing domain. Index Terms--Data mining, association rules, unexpectedness, minimality.