About

Shawn Curley is a faculty member associated with the Information & Decision Sciences department at the Carlson School of Management. He holds the position of Curtis L. Carlson Chair Professor in Business Analytics and Information Systems and serves as the Academic Director of the Carlson Analytics Lab. His expertise encompasses data analysis techniques, predictive analytics, programming, data engineering, and machine learning methods. As part of the Analytics for Good Institute, he contributes to initiatives that involve real-world data analysis projects, collaborating with graduate students and partner organizations to apply analytical techniques to business problems. His work supports the educational mission of the Carlson School by fostering the development of emerging data science professionals and promoting impactful analytics research.

Research topics

• Computer Science
• Advertising
• Psychology
• Business
• Mathematics
• Information Retrieval
• Marketing
• Telecommunications
• Social psychology
• Statistics

Selected publications

• Recommending Composite Items Using Multi-Level Preference Information: A Joint Interaction Modeling Approach

  ArXiv.org · 2026-01-26

  articleOpen accessSenior author

  With the advancement of machine learning and artificial intelligence technologies, recommender systems have been increasingly used across a vast variety of platforms to efficiently and effectively match users with items. As application contexts become more diverse and complex, there is a growing need for more sophisticated recommendation techniques. One example is the composite item (for example, fashion outfit) recommendation where multiple levels of user preference information might be available and relevant. In this study, we propose JIMA, a joint interaction modeling approach that uses a single model to take advantage of all data from different levels of granularity and incorporate interactions to learn the complex relationships among lower-order (atomic item) and higher-order (composite item) user preferences as well as domain expertise (e.g., on the stylistic fit). We comprehensively evaluate the proposed method and compare it with advanced baselines through multiple simulation studies as well as with real data in both offline and online settings. The results consistently demonstrate the superior performance of the proposed approach.

  PublisherOA PDF

• Recommending Composite Items Using Multi-Level Preference Information: A Joint Interaction Modeling Approach

  Open MIND · 2026-01-26

  preprintSenior author

  With the advancement of machine learning and artificial intelligence technologies, recommender systems have been increasingly used across a vast variety of platforms to efficiently and effectively match users with items. As application contexts become more diverse and complex, there is a growing need for more sophisticated recommendation techniques. One example is the composite item (for example, fashion outfit) recommendation where multiple levels of user preference information might be available and relevant. In this study, we propose JIMA, a joint interaction modeling approach that uses a single model to take advantage of all data from different levels of granularity and incorporate interactions to learn the complex relationships among lower-order (atomic item) and higher-order (composite item) user preferences as well as domain expertise (e.g., on the stylistic fit). We comprehensively evaluate the proposed method and compare it with advanced baselines through multiple simulation studies as well as with real data in both offline and online settings. The results consistently demonstrate the superior performance of the proposed approach.

  DOI

• Improving Business Performance in New Value Chains Using Standardisation and Injection Efficiency Performance Measures

  2025-09-02

  articleSenior author

  Abstract This paper demonstrates how applying ISO/TS 3250 for calculation and reporting Injection Efficiency (IE) can add value not only to the upstream oil and gas business category but also to new value chains such as carbon capture and storage (CCS). Utilizing performance measures is a well-established method for tracking and improving operating performance. Production efficiency (PE) is a term often used by operators for historic production availability in the operating phase. Requirements and guidance for PE and IE standardization are given in ISO/TS 3250:2021 "Petroleum, petrochemical and natural gas industries – Calculation and reporting production efficiency in the operating phase." Such standardization is important as it enables good quality benchmarking and more importantly, enables knowledge sharing and well-founded performance improvements to be identified and implemented within and between individual operators. The focus within the upstream sector is typically on Production Efficiency. For many fields, water injection, or gas injection are critical from both an environmental and reservoir pressure maintenance perspective. Although the impact is typically deferred, Injection Efficiency will ultimately impact Production Efficiency, production volumes, revenues, reserves, environmental performance, as well as other important performance measures within a field. In comparison to the traditional oil & gas upstream business category, the relationship between Injection Efficiency, injection volumes and revenues is more immediate and will therefore have a greater focus in the new value chains such as carbon capture and storage. Injection down time will represent a direct loss of income for the storage operator as well as its customers facing possible GHG emission costs due to venting. Having a consistent and standard approach for Injection Efficiency monitoring and applying this for business improvement should help ensure commercially attractive injection and storage projects. This paper discusses how the methodology and framework within ISO/TS 3250 can be applied for value chains that include injection and storage. ISO/TS 3250 was published August 2021 with a focus on Production Efficiency supplemented by Injection Efficiency measures. With the increased need for and focus on new value chains such as CCS, the implementation of standardized injection efficiency measures will significantly aid benchmarking and ultimately lead to improved business performance.

  PublisherDOI

• Forecasting Goal-Oriented Activities: A Deep Learning Approach Leveraging Cross-Activity Relationships

  SSRN Electronic Journal · 2023-01-01

  articleOpen accessSenior author
  PublisherDOI

• Persistence of Recommender-Induced Biases in Consumer Preference Ratings

  SSRN Electronic Journal · 2023-01-01

  articleOpen access
  PublisherDOI

• The Effects of Digitally Delivered Nudges in a Corporate Wellness Program

  Journal of the Association for Information Systems · 2023-01-01 · 5 citations

  article

  We investigate how two digitally delivered nudges, namely light social support (nonverbal cues such as kudos or likes) and motivational messaging, affect employees’ self-reported physical activity in an online, corporate wellness program. Within this unique field setting, using data from several years, we found evidence that both types of nudges provide benefits beyond the effect of cash incentives. However, the effects vary by individual, depending on whether the employee is actively engaging in physical activity, and by time, depending on how long the employee has been in the wellness program. We found light social support to be less effective over time, while motivational messages were found to be more effective with the duration in the program and generally more effective for physically inactive users. Our findings have implications for the design of wellness systems, suggesting different approaches depending on an employee’s current activity level and tenure in the program

  PublisherDOI

• Significance of Task Significance in Online Marketplaces for Work

  Proceedings of the ... Annual Hawaii International Conference on System Sciences/Proceedings of the Annual Hawaii International Conference on System Sciences · 2022-01-01

  articleOpen accessSenior authorCorresponding

  Online marketplaces for work like Amazon Mechanical Turk facilitate the sourcing of low expertise tasks in a fast and cost effective way. In this study, we explore the impact of task significance on work quality by informing workers of the purpose of the task and who benefits from it. Results from a laboratory experiment and a field experiment showed that perceived task significance improved work quality, but only for participants who recalled the purpose statement. In contrast, increasing monetary payment by 50% had no impact on work quality. A majority of participants who received the purpose statement were not able to recall it. Further analysis showed worker attributes such as English ability and personality traits influenced the likelihood of recall whereas rich media format had no effects. Overall, our work highlights the promise of task significance as a way to motivate online workers and the challenge of promoting task significance online.

  PublisherDOI

• Effects of Personalized Recommendations Versus Aggregate Ratings on Post-Consumption Preference Responses

  MIS Quarterly · 2022 · 24 citations

  • Computer Science
  • Information Retrieval
  • Psychology

  Online retailers use product ratings to signal quality and help consumers identify products for purchase. These ratings commonly take the form of either non-personalized, aggregate product ratings (i.e., the average rating a product received from a number of consumers such as “the average rating is 4.5/5 based on 100 reviews”), or personalized predicted preference ratings for a product (i.e., recommender-system-generated predictions for a consumer’s rating of a product such as “we think you’d rate this product 4.5/5”). Ratings in either format can provide decision aid to the consumer, but the two formats convey different types of product quality information and operate with different psychological mechanisms. Prior research has indicated that each recommendation type can significantly affect consumer’s post-experience preference ratings, constituting a judgmental bias, but has not compared the effects of these two common product-rating formats. Using a laboratory experiment, we show that aggregate ratings and personalized recommendations create similar biases on post-experience preference ratings when shown separately. Shown together, there is no cumulative increase in the effect. Instead, personalized recommendations tend to dominate. Our findings can help retailers determine how to use these different types of product ratings to most effectively serve their customers. Additionally, these results help to educate the consumer on how product-rating displays influence their stated preferences.

  PublisherDOI

• Recommender systems, ground truth, and preference pollution

  AI Magazine · 2022-06-01 · 6 citations

  articleOpen access

  Abstract Interactions between individuals and recommender systems can be viewed as a continuous feedback loop, consisting of pre‐consumption and post‐consumption phases. Pre‐consumption, systems provide recommendations that are typically based on predictions of user preferences. They represent a valuable service for both providers and users as decision aids. After item consumption, the user provides post‐consumption feedback (e.g., a preference rating) to the system, often used to improve the system's subsequent recommendations, completing the feedback loop. There is a growing understanding that this feedback loop can be a significant source of unintended consequences, introducing decision‐making biases that can affect the quality of the “ground truth” preference data, which serves as the key input to modern recommender systems. This paper highlights two forms of bias that recommender systems inherently inflict on the “ground truth” preference data collected from users after item consumption: non‐representativeness of such preference data and so‐called “preference pollution,” which denotes an unintended relationship between system recommendations and the user's post‐consumption preference ratings. We provide an overview of these issues and their importance for the design and application of next‐generation recommendation systems, including directions for future research.

  PublisherOA PDFDOI

• Effects of Personalized and Aggregate Top-N Recommendation Lists on User Preference Ratings

  ACM Transactions on Information Systems · 2021-01-14 · 9 citations

  article

  Prior research has shown a robust effect of personalized product recommendations on user preference judgments for items. Specifically, the display of system-predicted preference ratings as item recommendations has been shown in multiple studies to bias users’ preference ratings after item consumption in the direction of the predicted rating. Top-N lists represent another common approach for presenting item recommendations in recommender systems. Through three controlled laboratory experiments, we show that top-N lists do not induce a discernible bias in user preference judgments. This result is robust, holding for both lists of personalized item recommendations and lists of items that are top-rated based on averages of aggregate user ratings. Adding numerical ratings to the list items does generate a bias, consistent with earlier studies. Thus, in contexts where preference biases are of concern to an online retailer or platform, top-N lists, without numerical predicted ratings, would be a promising format for displaying item recommendations.

  PublisherDOI

Recent grants

• Strategies, Performance, and Acceptability in the Dynamic, Competitive Environment of Continuous Combinatorial Auctions

  NSF · $394k · 2010–2014

Frequent coauthors

• Gediminas Adomavičius

  42 shared

• Jesse Bockstedt

  Emory University

  21 shared

• Philip G. Benson

  21 shared

• Jingjing Zhang

  Nanjing Forestry University

  13 shared

• Glenn J. Browne

  Texas Tech University

  10 shared

• J. Frank Yates

  10 shared

• Pallab Sanyal

  George Mason University

  8 shared

• Gerald F. Smith

  8 shared