Tag Archives: machine learning

How Does AI-Generated Voice Affect Online Content Creation? Evidence from TikTok

Zhang, Xiaoke, Mi Zhou, Gene Moo Lee (2022) How Does AI-Generated Voice Affect Online Content Creation? Evidence from TikTokWork-in-progress.

Video is one of the fastest-growing online services offered to consumers. A growing number of people today are participating in video creation and consumption in the digital economy. We study whether and how AI-generated voice affects users’ routine efforts and creative efforts in online video creation. Using a unique dataset of 2,617 creators and 273,244 videos collected from TikTok over a 25-week period, we first exploit deep learning models to detect the adoption of AI-generated voice from massive video data. Then we estimate its treatment effect on creators and viewers using a difference-in-differences model coupled with propensity score matching. We find that AI-generated voice increases creators’ routine effort and creative effort in the short term. While it has a long-lasting effect on improving the efficiency of video creation, AI-generated voice cannot consistently motivate creators to include more information in videos, and might even be detrimental to their creative effort in the long term. Our study provides the first empirical evidence on how AI tools reshape video content creation patterns on online platforms, which carries important managerial implications for individual creators, platforms, and policymakers in the digital economy.

Ideas are Easy but Execution is Everything: Measuring the Impact of Stated AI Strategies and Capability on Firm Innovation Performance

Lee, Myunghwan, Gene Moo Lee (2022) “Ideas are Easy but Execution is Everything: Measuring the Impact of Stated AI Strategies and Capability on Firm Innovation Performance”Work-in-Progress.

Contrary to the promise that AI will transform various industries, there are conflicting views on the impact of AI on firm performance. We argue that existing AI capability measures have two major limitations, limiting our understanding of the impact of AI in business. First, existing measures on AI capability do not distinguish between stated strategies and actual AI implementations. To distinguish stated AI strategy and actual AI capability, we collect various AI-related data sources, including AI conferences (e.g., NeurIPS, ICML, ICLR), patent filings (USPTO), inter-firm transactions related to AI adoption (FactSet), and AI strategies stated in 10-K annual reports. Second, while prior studies identified successful AI implementation factors (e.g., data integrity and intelligence augmentation) in a general context, little is known about the relationship between AI capabilities and in-depth innovation performance. We draw on the neo-institutional theory to articulate the firm-level AI strategies and construct a fine-grained AI capability measure that captures the unique characteristics of AI-strategy. Using our newly proposed AI capability measure and a novel dataset, we will study the impact of AI on firm innovation, contributing to the nascent literature on managing AI.

Do Incentivized Reviews Poison the Well? Evidence from a Natural Experiment at Amazon.com

Park, Jaecheol, Arslan Aziz, Gene Moo Lee. “Do Incentivized Reviews Poison the Well? Evidence from a Natural Experiment at Amazon.comWorking Paper.

  • Presentations: UBC (2021), KrAIS (2021), WISE (2021), PACIS (2022), SCECR (2022), BU Platform (2022)

The rapid growth in e-commerce has led to a concomitant increase in consumers’ reliance on digital word-of-mouth to inform their choices. As such, there is an increasing incentive for sellers to solicit reviews for their products. Recent studies have examined the direct effect of receiving incentives or introducing incentive policy on review writing behavior. However, since incentivized reviews are often only a small proportion of the overall reviews on a platform, it is important to understand whether their presence on the platform has spillover effects on the unincentivized reviews which are often in the majority. Using the state-of-the-art language model, Bidirectional Encoder Representations from Transformers (BERT) to identify incentivized reviews, a document embedding method, Doc2Vec to create matched pairs of Amazon and non-Amazon branded products, and a natural experiment caused by a policy change on Amazon.com in October 2016, we conduct a difference-in-differences analysis to identify the spillover effects of banning incentivized reviews on unincentivized reviews. Our results suggest that there are positive spillover effects of the ban on the review sentiment, length, helpfulness, and frequency, suggesting that the policy stimulates more reviews in the short-run and more positive, lengthy, and helpful reviews in the long run. Thus, we find that the presence of incentivized reviews on the platform poisons the well of reviews for unincentivized reviews.

Learning Faces to Predict Matching Probability in an Online Dating Market

Kwon, Soonjae, Sung-Hyuk Park, Gene Moo Lee, Dongwon Lee. “Learning Faces to Predict Matching Probability in an Online Dating Market”. Working Paper.

  • Presentations: DS (2021), AIMLBA (2021), WITS (2021), ICIS (2022)
  • Based on an industry collaboration

With the increasing use of online matching platforms, predicting matching probability between users is crucial for efficient market design. Although previous studies have constructed various visual features to predict matching probability, facial features, which are important in online matching, have not been widely used. We find that deep learning-enabled facial features can significantly enhance the prediction accuracy of a user’s partner preferences from the individual rating prediction analysis in an online dating market. We also build prediction models for each gender and use prior theories to explain different contributing factors of the models. Furthermore, we propose a novel method to visually interpret facial features using the generative adversarial network (GAN). Our work contributes to the literature by providing a framework to develop and interpret facial features to investigate underlying mechanisms in online matching markets. Moreover, matching platforms can predict matching probability more accurately for better market design and recommender systems.

My thoughts on AI, Big Data, and IS Research

Last update: June 10th, 2021

Recently, I had a chance to share my thoughts on how Big Data Analytics and AI will impact Information Systems (IS) research. Thanks to ever-growing datasets (public and proprietary) and powerful computational resources (cloud API, open-source projects), AI and Big Data will be important in IS research in the foreseeable future. If you are an aspiring IS researcher, I believe that you should be able to embrace this and take advantage of this.

First, AI and Big Data are powerful “tools” for IS research. It could be intimidating to see all the fancy new AI techniques. But they are just tools to analyze your data. You don’t need to reinvent the wheel to use them. There are many open-source projects in Python and R that you can use to analyze your data. Also, many cloud services (e.g., Amazon Rekognition, Google Cloud ML, Microsoft Azure ML) allow you to use pre-trained AI models at a modest cost (that your professors can afford). What you need is some working knowledge in programming languages like Python and R. And a high-level understanding of the idea behind algorithms.

Don’t shy away from hands-on programming. Using AI and Big Data tools may not be a competitive advantage in the long run because of the democratization of AI tools. However, I believe it will be the new baseline. So you need to have it in your research toolbox. Specifically, I believe that IS researchers should have a working knowledge of Python/R programming and Linux environment. I recommend these online courses: Data ScienceMachine LearningLinuxSQL, and NoSQL.

Second, AI and Big Data Analytics are creating a lot of interesting new “phenomenon” in personal lives, firms, and societies. How AI and robots will be adopted in the workplace and how that will affect the labor market? Are we losing our jobs? Or can we improve our productivity with AI tools? How AI will be used in professional services by the experts? What are the unintended consequences (such as biases, security, privacy, misinformation) of AI adoptions in the organization and society? And how can we mitigate such issues? There are so many new and interesting research questions.

In order to conduct relevant research, I think that IS researchers should closely follow the emerging technologies. Again, it could be hard to keep up with all the advances. I try to keep up to date by reading industry reports (from McKinsey and Deloitte) and listening to many podcasts (e.g., Freakonomics Radio, a16 Podcasts by Andreessen Horowitz, Lex Fridman Podcast, Stanford’s Entrepreneurial Thought Leaders, HBR’s Exponential View by Azeem Azhar).

I hope this post may help new IS researchers shape their research strategies. I will try to keep updating this post. Cheers!

 

 

IS / Marketing Papers on Visual Data Analytics (Image, Video)

Last update: June 9, 2022

With the advent of social media and mobile platforms, visual data are becoming the first citizen in big data analytics research. Compared to textual data that require significant cognitive efforts to comprehend, visual data (such as images and videos) can easily convey the message from the content creator to the general audience. To conduct large-scale studies on such data types, researchers need to use machine learning and computer vision approaches. In this post, I am trying to organize studies in Information Systems, Marketing, and other management disciplines that leverage large-scale analysis of image and video datasets. The papers are ordered randomly:

  1. Bharadwaj, N., Ballings, M., Naik, P. A., Moore, M, Arat, M. M. (2022) “A New Livestream Retail Analytics Framework to Assess the Sales Impact of Emotional Displays,” Journal of Marketing, 86(1): 24-47.
  2. Chen, Z., Liu, Y.-J., Meng, J., Wang, Z. (2022) “What’s in a Face? An Experiment on Facial Information and Loan-Approval Decision“, Management Science, forthcoming.
  3. Lu, T., Wang, A., Yuan, X., Zhang, X. (2020) “Visual Distortion Bias in Consumer Choices,” Management Science, forthcoming.
  4. Zhou, M., Chen, G. H., Ferreira, P., Smith, M. D. (2021) “Consumer Behavior in the Online Classroom: Using Video Analytics and Machine Learning to Understand the Consumption of Video Courseware,” Journal of Marketing Research 58(6): 1079-1100.
  5. Zhang, S., Lee, D., Singh, P. V., Srinivasan, K. (2021) “What Makes a Good Image? Airbnb Demand Analytics Leveraging Interpretable Image Features,” Management Science, forthcoming.
  6. Gunarathne, P., Rui, H., Seidmann, A. (2021) “Racial Bias in Customer Service: Evidence from Twitter,” Information Systems Research 33(1): 43-54.
  7. Shin, D., He, S., Lee, G. M., Whinston, A. B., Cetintas, S., Lee, K.-C. (2020) “Enhancing Social Media Analysis with Visual Data Analytics: A Deep Learning Approach,” MIS Quarterly 44(4): 1459-1492[Details]
  8. Li, Y., Xie, Y. (2020) “Is a Picture Worth a Thousand Words? An Empirical Study of Image Content and Social Media Engagement,” Journal of Marketing Research 57(2): 1-19.
  9. Zhang, Q., Wang, W., Chen, Y. (2020) “Frontiers: In-Consumption Social Listening with Moment-to-Moment Unstructured Data: The Case of Movie Appreciation and Live comments,” Marketing Science 39(2).
  10. Liu, L., Dzyabura, D., Mizik, N. (2020) “Visual Listening In: Extracting Brand Image Portrayed on Social Media,Marketing Science 39(4): 669-686.
  11. Peng, L., Cui, G., Chung, Y., Zheng, W. (2020) “The Faces of Success: Beauty and Ugliness Premiums in E-Commerce Platforms,” Journal of Marketing 84(4): 67-85.
  12. Liu, X., Zhang, B., Susarla, A., Padman, R. (2020) “Go to YouTube and Call Me in the Morning: Use of Social Media for Chronic Conditions,” MIS Quarterly 44(1b): 257-283.
  13. Zhao, K., Hu, Y., Hong, Y., Westland, J. C. (2020) “Understanding Characteristics of Popular Streamers in Live Streaming Platforms: Evidence from Twitch.tv,” Journal of the Association for Information Systems, Forthcoming.
  14. Ordenes, F. V., Zhang, S. (2019) “From words to pixels: Text and image mining methods for service research,” Journal of Service Management 30(5): 593-620.
  15. Wang, Q., Li, B., Singh, P. V. (2018) “Copycats vs. Original Mobile Apps: A Machine Learning Copycat-Detection Method and Empirical Analysis,” Information Systems Research 29(2): 273-291.
  16. Lu, S., Xiao, L., Ding, M. (2016) “A Video-Based Automated Recommender (VAR) System for Garments,” Marketing Science 35(3): 484-510.
  17. Xiao, L., Ding, M. (2014) “Just the Faces: Exploring the Effects of Facial Features in Print Advertising,” Marketing Science 33(3), 315-461.
  18. Suh, K.-S., Kim, H., Suh, E. K. (2011) “What If Your Avatar Looks Like You? Dual-Congruity Perspectives for Avatar Use,” MIS Quarterly 35(3), 711-729.
  19. Todorov, A., Porter, J. M. (2014) “Misleading First Impressions: Different for Different Facial Images of the Same Person“, Psychological Science 25(7): 1404-1417.
  20. Todorov, A., Madnisodza, A. N., Goren, A., Hall, C. C. (2005) “Inferences of Competence from Faces Predict Election Outcomes“, Science 308(5728): 1623-1626.
  21. Mueller. U., Mazur, A. (1996) “Facial Dominance of West Point Cadets as a Predictor of Later Military Rank“, Social Forces 74(3): 823-850.
  22. Lee, H, Nam, K. “When Machine Vision Meets Human Fashion: Effects of Human Intervention on the Efficiency of CNN-Driven Recommender Systems in Online Fashion Retail”, Working Paper.
  23. Lysyhakov M, Viswanathan S (2021) “Threatened by AI: Analyzing users’ responses to the introduction of AI in a crowd-sourcing,” Working Paper.
  24. Park, S., Lee, G. M., Shin, D., Han, S.-P. (2020) “Targeting Pre-Roll Ads using Video Analytics,” Working Paper.
  25. Choi, A., Ramaprasad, J., So, H. (2021) Does Authenticity of Influencers Matter? Examining the Impact on Purchase Decisions, Working Paper.
  26. Park, J., Kim, J., Cho, D., Lee, B. Pitching in Character: The Role of Video Pitch’s Personality Style in Online Crowdsourcing, Working Paper.
  27. Yang, J., Zhang, J., Zhang Y. (2021) First Law of Motion: Influencer Video Advertising on TikTok, Working Paper.
  28. Davila, A., Guasch (2021) Manager’s Body Expansiveness, Investor Perceptions, and Firm Forecast Errors and Valuation, Working Paper.
  29. Peng, L., Teoh, S. H., Wang, U., Yan, J. (2021) Face Value: Trait Inference, Performance Characteristics, and Market Outcomes for Financial Analysts, Working Paper.
  30. Zhang, S., Friedman, E., Zhang, X., Srinivasan, K., Dhar, R. (2020) Serving with a Smile on Airbnb: Analyzing the Economic Returns and Behavioral Underpinnings of the Host’s Smile,” Working Paper.
  31. Park, K., Lee, S., Tan, Y. (2020) “What Makes Online Review Videos Helpful? Evidence from Product Review Videos on YouTube,” UW Working Paper.
  32. Doosti, S., Lee, S., Tan, Y. (2020) “Social Media Sponsorship: Metrics for Finding the Right Content Creator-Sponsor Matches,” UW Working Paper.
  33. Koh, B., Cui, F. (2020) “Give a Gist: The Impact of Thumbnails on the View-Through of Videos,” KU Working Paper.
  34. Hou J.R., Zhang J., Zhang K. (2018) Can title images predict the emotions and the performance of crowdfunding projects? Workshop on e-Business.

Trustworthy Face? The Effect and Drivers of Comprehensive Trust in Online Job Market Platform

Kwon, Jun Bum, Donghyuk Shin, Gene Moo Lee, Jake An, Sam Hwang (2020) “Trustworthy Face? The Effect and Drivers of Comprehensive Trust in Online Job Market Platform”. Work-in-progress.

The abstract will appear here.

Robots Serve Humans: Does AI Robot Adoption Enhance Operational Efficiency and Customer Experience?

Lee, Myunghwan, Gene Moo Lee, Donghyuk Shin, Sang-Pil Han (2022) “Robots Serve Humans? Understanding the Economic and Societal Impacts of AI Robots in the Service IndustryWorking Paper.

  • Presented at WITS (2020), KrAIS (2020), UBC (2021), DS (2022)
  • Research assistants: Raymond Situ, Gallant Tang

Service providers, such as restaurants, have been adopting various robotics technologies to improve operational efficiency and increase customer satisfaction. AI Robotics technologies bring new restaurant experiences to customers by taking orders, cooking, and serving. While the impact of industrial robots has been well documented in the literature, little is known about the impact of customer-facing service robot adoption. To fill this gap, this work-in-progress study aims to analyze the impact of service robot adoption on restaurant service quality using 4,610 restaurants and their online customer reviews. We analyzed the treated effect of robot adoption using a difference-in-differences approach with propensity score and exact matching. Estimation results show that restaurant robot adoption has a positive impact on customer satisfaction, specifically on perceived service quality. This study provides both academic and practical implications on emerging AI robotics techniques.

Corporate Social Network Analysis: A Deep Learning Approach

Cao, Rui, Gene Moo Lee, Hasan Cavusoglu. “Corporate Social Network Analysis: A Deep Learning Approach,” Working Paper.

Identifying inter-firm relationships is critical in understanding the industry landscape. However, due to the dynamic nature of such relationships, it is challenging to capture corporate social networks in a scalable and timely manner. To address this issue, this research develops a framework to build corporate social network representations by applying natural language processing (NLP) techniques on a corpus of 10-K filings, describing the reporting firms’ perceived relationships with other firms. Our framework uses named-entity recognition (NER) to locate the corporate names in the text, topic modeling to identify types of relationships included, and BERT to predict the type of relationship described in each sentence. To show the value of the network measures created by the proposed framework, we conduct two empirical analyses to see their impacts on firm performance. The first study shows that competition relationship and in-degree measurements on all relationship types have prediction power in estimating future earnings. The second study focuses on the difference between individual perspectives in an inter-firm social network. Such a difference is measured by the direction of mentions and is an indicator of a firm’s success in network governance. Receiving more mentions from other firms is a positive signal to network governance and it shows a significant positive correlation with firm performance next year.

When Does Congruence Matter for Pre-roll Video Ads? The Effect of Multimodal, Ad-Content Congruence on the Ad Completion

Park, Sungho, Gene Moo Lee, Donghyuk Shin, Sang-Pil Han. “When Does Congruence Matter for Pre-roll Video Ads? The Effect of Multimodal, Ad-Content Congruence on the Ad Completion, Under Review [Submitted: June 27, 2022]

  • Previous title: Targeting Pre-Roll Ads using Video Analytics
  • Funded by Sauder Exploratory Research Grant 2020
  • Presented at Southern Methodist University (2020), University of Washington (2020), INFORMS (2020), AIMLBA (2020), WITS (2020), HKUST (2021), Maryland (2021), American University (2021), National University of Singapore (2021)
  • Research assistants: Raymond Situ, Miguel Valarao

Pre-roll video ads are gaining industry traction because the audience may be willing to watch an ad for a few seconds, if not the entire ad, before the desired content video is shown. Conversely, a popular skippable type of pre-roll video ads, which enables viewers to skip an ad in a few seconds, creates opportunity costs for advertisers and online video platforms when the ad is skipped. Against this backdrop, we employ a video analytics framework to extract multimodal features from ad and content videos, including auditory signals and thematic visual information, and probe into the effect of ad-content congruence at each modality using a random matching experiment conducted by a major video advertising platform. The present study challenges the widely held view that ads that match content are more likely to be viewed than those that do not, and investigates the conditions under which congruence may or may not work. Our results indicate that non-thematic auditory signal congruence between the ad and content is essential in explaining viewers’ ad completion, while thematic visual congruence is only effective if the viewer has sufficient attentional and cognitive capacity to recognize such congruence. The findings suggest that thematic videos demand more cognitive processing power than auditory signals for viewers to perceive ad-content congruence, leading to decreased ad viewing. Overall, these findings have significant theoretical and practical implications for understanding whether and when viewers construct congruence in the context of pre-roll video ads and how advertisers might target their pre-roll video ads successfully.