Research Interests

Quantitative modeling of social influence and shared consumption, Firm and user generated contents, Multi-methods, Bayesian statistical inference, Machine learning, Business networks, Entertainment industry

Research

Ph.D. Office during Michigan years
The two main research streams that I am broadly interested in are (1) Quantitative modeling of firm- and user-generated contents and (2) Quantitative modeling of social influence in business and consumer settings. I use a wide range of quantitative/qualitative methods for my research, including image processing, deep learning, running experiments and collecting physiological measures.

Please see below for the latest selected list of published and working papers. For the full list, please refer to my CV.

​Last update: Aug. 2025
​​^ : a supervised student or a student collaborator (MS/PhD level); ​* : equal contribution

Quantitative modeling of firm and user generated contents

"The Moral Significance of Aesthetic Quality in Nature Imagery" [Link]
Kim, Eunsoo, Julia Lee Cunningham and Anocha Aribarg (2022)
​​Psychological Science, 33(9), 1372-1385.     ​
  • Nominated and covered in the university-wide NTU research magazine, Pushing Frontiers, Issue 22 (p. 7, Discoveries section)  
    ​“AI and Marketing: People Care When Faced With Nature's Beauty”
  • Using deep learning and experimental studies, we show the causal effect of visual aesthetics on social media engagement and moral concern, which is explained by self-transcendent emotions (awe and inspiration) and purity associated with an image.
​"Does Topic Consistency Matter? A Study of Critic and User Reviews in the Movie Industry" [Link]
​Kim, Eunsoo, MengQi (Annie) Ding^, Xin (Shane) Wang and Shijie Lu (2023)
Journal of Marketing, 87(3), 428-450.
  • ​Covered in the inaugural issue of NBS Research Magazine, Issue 1, 2023
  • Covered in NTU/NBS Newsletters: Thought Leadership & NBS Highlights
  • This research shows the topic consistency, or the degree of content overlap between critic and user review content, increases movie demand, and suggest a potential mechanism through increased memorability.
“Movie Box-Office Prediction Using a Self-Supervised Learning Transformer Encoder" [Link]
Chao, Qin^, ​Eunsoo Kim and Boyang Li (2023), Proceedings of IEEE ICME, 1535-1540. 

"Copycat vs. Original: Multi-modal Pretraining and Variable Importance in Box Office Prediction"
Chao, Qin^, Eunsoo Kim and Boyang Li (accepted, IEEE Transactions on Multimedia)

​“# Gifted: The Impact of Influencer Gifting Strategy in Influencer Marketing”
Wang, Ruoding^, Eunsoo Kim and Xinlong Li (accepted, International Journal of Advertising) 

“The Body Speaks: The Effects of Machine-Extracted Body Pose in Image Contents” 
Lee, Yeonsoo^*, Eunsoo Kim* and Hye-jin Kim* 

Quantitative modeling of social influence

​"Herding in NFT Auction: The Role of Visual Complexity and Familiarity" [Link]
​Xie, Peiwen^, Eunsoo Kim, Shun Yin Lam and Sadat Reza (accepted, International Journal of Research in Marketing)

"Avatar Humanness and Psychological Closeness in Virtual Reality" [Link]
​Chae, Rebecca*, Hyojin Lee* and Eunsoo Kim (2025)
Psychology and Marketing, 42(4), 1132-1145.

​​"Personal versus Business Factors in Network Formation:  An Empirical Analysis of the Network Marketing Industry" 
​​Kim, Eunsoo, Puneet Manchanda and Xu Zhang  (working paper, University of Michigan) 
​
"The Impact of Network and Spatial Embeddedness on Salespeople Inactivity in Direct Selling Organizations" 
​​Kim, Eunsoo, Puneet Manchanda  (working paper, University of Michigan) 
"How Shared Consumption Enhances Entertainment Experiences" 
​​​Kim, Eunsoo, Anocha Aribarg and Natasha Zhang Foutz 
- ​Featured in Marketing Science Institute ( MSI ) Working Paper Series, Report No. 16 - 126