MIS Distinguished Speaker Series

Temple University

You are here: Home / Archives for Jing Gong

Jing Gong

October 5 – Anand Gopal to Present “A for Effort? Using the Crowd to Identify Moral Hazard in NYC Restaurant Hygiene Inspections”

By

Department of Management Information Systems and Data Science Institute

A for Effort? Using the Crowd to Identify Moral Hazard in NYC Restaurant Hygiene Inspections

by

Anandasivam Gopal

Dean’s Professor of Information Systems

Robert H. Smith School of Business, University of Maryland

Friday, October 5, 2018

10:30 AM – noon

Fred Fox Boardroom (Alter 378)

Abstract

From an upset stomach to a life-threatening foodborne illness, getting sick is all too common after eating in restaurants. While health inspection programs are designed to protect consumers, such inspections typically occur at wide intervals of time, allowing restaurant hygiene to remain unmonitored in the interim periods. Information provided in online reviews may be effectively used in these interim periods to gauge restaurant hygiene. In this paper, we provide evidence for how information from online reviews of restaurants can be effectively used to identify cases of hygiene violations in restaurants, even after the restaurant has been inspected and certified. We use data from restaurant hygiene inspections in New York City from the launch of an inspection program from 2010 to 2016, and combine this data with online reviews for the same set of restaurants. Using supervised machine learning techniques, we then create a hygiene dictionary specifically crafted to identify hygiene-related concerns, and use it to identify systematic instances of moral hazard, wherein restaurants with positive hygiene inspection scores are seen to regress in their hygiene maintenance within 90 days of receiving the inspection scores. To the extent that social media provides some visibility into the hygiene practices of restaurants, we argue that the effects of information asymmetry that lead to moral hazard may be partially mitigated in this context. Based on our work, we also provide strategies for how cities and policy-makers may design effective restaurant inspection programs, through a combination of traditional inspections and the appropriate use of social media.

September 28 – Mohammad Saifur Rahman to Present “Where You Live Matters: The Impact of Local Financial Market Competition in Managing Online Peer-To-Peer Loans”

By

Where You Live Matters: The Impact of Local Financial Market Competition in Managing Online Peer-To-Peer Loans

by

Mohammad Saifur Rahman

Associate Professor of Management

Krannert School of Management, Purdue University

Friday, September 28, 2018

10:30 AM – noon

Speakman Hall Suite 200

 

Abstract

Internet related technologies have fundamentally changed many industries, and, in the age of financial technology (FinTech), a question that is being widely discussed is whether the local financial market structure still matters. Unlike traditional retail financial institutions, which are predominantly territorial, FinTech products — in particular, peer-to-peer (P2P) lending platforms — provide equal access to funds to borrowers from across the country, removing any typical geographic restrictions in borrowing options. However, if P2P lending platforms are not immune to competition from local financial institutions and borrowers ultimately gain from the strategic interactions between the local financial institutions and P2P platforms, where a borrower lives might continue to matter! Consequently, we study the impact of local financial market structure on borrowers’ personal loan management decisions — to prepay or to default — on the two leading P2P lending platforms, Lending Club and Prosper. We find consistently, across the two platforms, that an online borrower from a more competitive market is more likely to prepay and less likely to default. Additionally, this study offers novel insights regarding the extent and nature of the substitution between traditional financial institutions and their online, potentially disruptive, alternatives. Also, we utilize machine learning techniques that capitalize on the rich granularity of the data set to create a pseudo-experimental design and further validate the underlying mechanism behind our results. Going beyond P2P lending, these findings suggest that borrowers benefit disproportionately, based on their geographic location, from local lending institutions. We discuss managerial, practical, and policy implications for the burgeoning P2P lending industry as well as other crowd-based markets.

September 14 – Andrew Burton-Jones to Present “Evaluating Digital Transformation in Healthcare: An Institutional Theory Perspective”

By

Evaluating Digital Transformation in Healthcare: An Institutional Theory Perspective

by

Andrew Burton-Jones

Professor of Business Information Systems

UQ Business School, University of Queensland

Friday, September 14, 2018

10:30 AM – noon

Speakman Hall Suite 200

 

Abstract

Like many other industries, the global health sector is engaged in significant digital transformation. Given the major investments, and the major consequences for numerous stakeholders, evaluations are important. However, many studies have critiqued both the quality of evaluations and the quality of evaluation research. The persistent lack of progress in this field has led researchers to ask deeper questions about what is actually occurring when teams attempt to measure the benefits of digital transformation. This translational research essay explores how Institutional Theory offers a useful lens for understanding the complexities of evaluation and provides insights for improving research and practice. In particular, we show how Institutional Theory can explain numerous behaviors observed in the literature and in our own case study. We also show how Institutional Theory can benefit from the insights observed in evaluation work. Motivated by these opportunities, we suggest a research agenda through which practitioners and researchers can improve work in this area.

Bio

Andrew Burton-Jones is a Professor of Business Information Systems at the UQ Business School, University of Queensland. He has a Bachelor of Commerce (Honours) and Masters of Information Systems from the University of Queensland and a Ph.D. from Georgia State University. He is a Senior Editor of MIS Quarterly has served on the Editorial Boards of MIS Quarterly, Information Systems Research, Journal of the Association for Information Systems, Information & Organization, and other journals. He has also served as Program Co-Chair for AMCIS and PACIS, and has received several awards for his research, teaching, and service. He conducts research on systems analysis and design, the effective use of information systems, and conceptual/methodological issues. Prior to his academic career, he was a senior consultant in a big-4 accounting/consulting firm.

May 4 – Alexander Tuzhilin to Present “Learning to Generate Indistinguishable Product Reviews”

By

Learning to Generate Indistinguishable Product Reviews

by

Alexander Tuzhilin

Professor of Information Systems and the Leonard N. Stern Professor of Business

NYU Stern School of Business

Friday, May 4, 2018

10:30 AM – noon

Speakman Hall Suite 200

 

Abstract

In this paper, we purpose a novel method called RevGAN to generate user reviews using a combination of Hierarchical AutoEncoder (hAE) and Conditional GAN (cGAN). We describe the proposed method and empirically demonstrate that it significantly outperforms several important benchmarks on the Amazon Review Dataset, and is also empirically indistinguishable from organic user reviews.

April 20 – Huseyin Cavusoglu to Present “Is Free Shipping Really Free? Strategic Implications of Membership-Based Free Shipping Programs of Online Marketplaces”

By

Is Free Shipping Really Free? Strategic Implications of Membership-Based Free Shipping Programs of Online Marketplaces

by

Huseyin Cavusoglu

Professor of Information Systems

Naveen Jindal School of Management, University of Texas at Dallas

Friday, April 20, 2018

10:30 AM – noon

Speakman Hall Suite 200

 

Abstract

We examine the membership-based free shipping (MFS) program offered by some online marketplaces in which a retail platform bears the shipping costs for purchases made by members that have paid an upfront fee, but non-members bear the shipping costs themselves. We show that the membership fee collected by the platform from members does not cover the cost of shipping products to members for their purchases during the membership period. While it may appear from this finding that the MFS program benefits members and hurts the platform, we show the MFS program actually benefits the platform when the shipping cost is less than a threshold value, which is increasing in the commission rate the platform earns from the third-party sellers. However, the gain from the MFS program is not necessarily decreasing in the shipping cost either. The MFS program always hurts non-members; it may also hurt even members. The demand enhancement, price increasing, and negative externality effects of the MFS program explain the results. Our findings imply that judging the success of the MFS program to either the platform or members solely based on the membership fee and shipping cost is misleading and that the MFS program is most attractive to the platform when the shipping cost is neither too low nor too high. Finally, the society can be worse off under the MFS program because the MFS program may stimulate demand from some low valuation and high misfit cost members who would not make a purchase in the absence of the MFS program, but the surplus enjoyed by these members is offset by the shipping cost borne by the platform.

April 6 – John D’Arcy to Present “Seeing the Forest and the Trees: A Meta-Analysis of the Antecedents to Information Security Policy Compliance”

By

Seeing the Forest and the Trees: A Meta-Analysis of the Antecedents to Information Security Policy Compliance

by

John D’Arcy

Associate Professor of MIS

Lerner College of Business and Economics, University of Delaware

 

Friday, April 6, 2018

10:30 AM – noon

Speakman Hall Suite 200

 

Abstract

A rich stream of research has identified numerous antecedents to employee compliance (and non-compliance) with information security policies. However, the number of competing theoretical perspectives and inconsistencies in the reported findings have hampered efforts to attain a clear understanding of what truly drives this behavior. To address this theoretical stalemate and build toward a consensus on the key antecedents of employees’ security policy compliance in different contexts, we conducted a meta-analysis of the relevant literature. Drawing on 84 quantitative studies focusing on security policy compliance, we classified 299 independent variables into 17 distinct categories and analyzed each category’s relationship with security policy compliance, including an analysis for possible domain-specific moderators. We augmented our meta-analytic assessment of the bivariate relationships between the independent variables and security policy compliance with a relative weight analysis that accounted for several construct intercorrelations. Collectively, our results suggest that much of the security policy compliance literature is plagued by suboptimal theoretical framing. Our findings can facilitate more refined theory-building efforts in this research domain and serve as a guide for practitioners to manage policy compliance initiatives.

March 16 – Amrit Tiwana to Present: “From Products to Platforms: Opportunities and Challenges for Practitioners”

By

From Products to Platforms: Opportunities and Challenges for Practitioners

by

Amrit Tiwana

P. George Benson Professor of Management Information Systems

Terry College of Business, University of Georgia

Friday, March 16, 2018

10:30 AM – 12:00 PM

Speakman Hall Suite 200

Abstract

What’s driving the morphing of products into platforms, why a product mindset falls flat in platform markets, how concepts absent in the vocabulary of product firms define their competitive dynamics, and hard questions product firms must ask themselves to make the transition. We explore these questions from a practitioner perspective based on fieldwork in a variety of firms in the US, Japan, and Europe.

Bio

Amrit Tiwana (people.terry.uga.edu/tiwana) is the P. George Benson Professor at the University of Georgia’s Terry College of Business. His research has been supported by organizations including Fuji Xerox, Fujitsu, IBM, Kansai Electric, Mitsubishi Electric, NEC, UPS, Mitsui, NTT Japan, SAP, Sumitomo Steel, Sony, Toshiba, and Hitachi. His two recent books were on platform competition (Morgan Kaufman, 2014) and on IT strategy (MIT Press, 2017). He serves or has served on the editorial boards of Information Systems Research (ISR), Strategic Management Journal (SMJ), Journal of Management Information Systems (JMIS), and MIS Quarterly.

Tagged With:

March 30 – H. Raghav Rao to Present “A Longitudinal Study of Unauthorized Access Attempts on Information Systems: The Role of Opportunity Contexts”

By

A Longitudinal Study of Unauthorized Access Attempts on Information Systems: The Role of Opportunity Contexts

by

H. Raghav Rao

AT&T Distinguished Chair in Infrastructure Assurance and Security
Department of Information Systems and Cyber Security
College of Business, University of Texas at San Antonio

Friday, March 30, 2018

10:30 AM – noon

Speakman Hall Suite 200

 

Abstract

This study investigates employees’ unauthorized access attempts on information systems (IS) applications in a financial institution and how opportunity contexts impact such attempts. By contextualizing multilevel criminal opportunity theory, we develop a model that considers both employee- and department-level opportunity contexts. At the employee level, we hypothesize that the number of IS apps an employee has legitimately accessed and the level of confidentiality of those apps, together with the time when and the location where the employee initiated the access, affect the likelihood of unauthorized attempts. At the department level, we hypothesize that department size moderates the impact of employee-level contextual variables on the likelihood of unauthorized attempts occurring. To test the hypotheses, we collected six months of access log data from an enterprise single sign-on system of a financial institution. We find the hypothesized main effects of all employee-level contextual variables are supported. In addition, department size reinforces the effects of off-hour access, off-site access, and their interaction term. Robustness analyses indicate that the results do not align with employees who do not know the systems well enough and may be making mistakes. We also discuss the theoretical and practical implications of the study.

February 23 – Giri Kumar Tayi to Present “Try-Before-You-Buy (TBYB): Online Retailing Strategy with Customer Self-Mending”

By

Try-Before-You-Buy (TBYB): Online Retailing Strategy with Customer Self-Mending

by

Giri Kumar Tayi

 Professor of Management Science and Information Systems
School of Business, SUNY Albany

Friday, February 23, 2018

10:30 AM – noon

Speakman Hall Suite 200

 

Abstract

In this study, we develop a parsimonious model to investigate the Try-Before-You-Buy (TBYB) strategy for an online retailer considering customer self-mending (CSM) behaviors. Our findings show that CSM behaviors have a great impact on the TBYB strategy. Without CSM, we find that the retailer offers no price discount under TBYB strategy. Nevertheless, TBYB strategy works as a Pareto improvement mechanism, even a “win-win” strategy for the society. With CSM, the results are quite different and thus interesting. First, we find the retailer’s offering of price discount in TBYB strategy is impacted by the product value, the return hassle cost as well as the size of the member segment. Second, we find the retailer’s adoption of TBYB strategy critically depends on the operation/return handling cost, product value, and the size of member segments. Specifically, the TBYB strategy is viable if: 1) the operation/return handling cost is low enough; or 2) the operation/return handling cost is relatively high, but the product value is large enough; or 3) although the operation/return handling cost is relatively high, and the product value is relatively small, the size of member segment is large enough. Third, we find the TBYB strategy reduces customer surplus on average, but it increases social welfare under some conditions. In addition, based on a generalized model, we find the price discount structure is additionally affected by the product mix and the number of products members keep in TBYB strategy. Further, we find that the TBYB strategy is also a “win-win” strategy in a competitive market setting. Such findings offer guidelines for online retailers that encounter problems related to product fit uncertainty to design appropriate TBYB strategies.

February 16 – Ravi Aron to Present “Is That A Sugar High or High Sugar? Ask The Algorithm!”

By

Is That A Sugar High or High Sugar? Ask The Algorithm!

by

Ravi Aron

Associate Professor, Information Systems
Carey Business School, Johns Hopkins University

Friday, February 16, 2018

10:30 AM – noon

Speakman Hall Suite 200

 

Abstract

We use methods based on machine learning to predict the transition of health state in chronic care patients. We collected data from a large, tertiary care, multi-specialty urban hospital system. In these hospitals about 2450 patients were in a diabetes case management program and from this cohort we obtained the EMRs of 1,687 patients. We use a two-stage machine learning algorithm that combines structured and unstructured data to predict the transitions in patients’ health states. We use a technique that we term ‘Differential Parsing’ where the output of the algorithm’s first stage prediction is used to identify those instances where there are key information signals in unstructured data. We then use LDA algorithm to extract decision rules from unstructured data which then enter as inputs for the second stage of the prediction. We construct four ML algorithms and contrast these with regression-based models. We find that the use of unstructured data improves the prediction accuracy of ML algorithms and they outperform regression-based models.