KKKarthik Natarajan Kannan

Karthik Natarajan Kannan

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Karthik Natarajan Kannan has a total of 49 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Unraveling the Impact: An Empirical Investigation of ChatGPT’s Exclusion from Stack Overflow

    by Sameer Borwankar, Warut Khern-Am-Nuai, Anastasiya Ghosh, Karthik Kannan

    This study explores the implications of a policy change announced by Stack Overflow, a popular question-and-answer platform, to prohibit the use of ChatGPT for generating questions or answers. To analyze data from Stack Overflow and Reddit, we employ the difference-in-differences (DID) empirical strategy to examine the impact of Stack Overflow’s policy. Our analysis focuses on the linguistic characteristics of content using natural language processing (NLP) techniques and voting features on both platforms. The results indicate that, after the restriction, the users on Stack Overflow exhibited a change in their answer generation, producing answers that exhibited more language complexity, positivity, and a larger number of words than the users on the AskProgramming subreddit. However, there was no discernible impact on the questions asked on Stack Overflow. Notably, the answers on Stack Overflow received more upvotes after the restriction, suggesting that the community perceived them to be of higher quality than answers on the subreddit. The frequency analysis reveals a noticeable decrease in the number of questions and answers posted by users on Stack Overflow following the restriction. Additionally, our supplementary analysis suggests a decrease in first-time contributors on Stack Overflow after the restriction. We validate our findings using an independent natural experiment leveraging Italy’s temporary nationwide ChatGPT ban, as well as a scenario-based experiment with 440 participants that provides direct evidence of compensatory knowledge signaling by contributors when AI-generated content is restricted. This study contributes to the existing literature by providing empirical evidence and insights into the decision to prohibit ChatGPT on a question-and-answer platform and provides practical implications to platform managers, informing them about the impact of the restriction on the change in answer content.

  • Open Access

    Information Design in Consumer-to-Consumer (C2C) Marketplaces: The Roles of Anonymity and Product Quality Uncertainty

    by Sameer Borwankar, Rajiv Mukherjee, Karthik Kannan

    Recently, there has been a notable increase in the prevalence of consumer-to-consumer (C2C) online marketplaces offering integration with social networks. Platforms such as eBay, Facebook Marketplace, Zhuanzhuan, and Poshmark have led such initiatives. In such marketplaces, the structure of social relationships revealed to platform participants can influence the choice of transacting partners as well as the terms of transactions. We study the role of different levels of information design policies on offers and acceptance decisions in a social commerce marketplace setting. We propose a novel information design policy, for C2C marketplaces, called partial anonymity in which the transacting parties are provided with limited social information about each other. Using a theoretical model and experimental data, we show that a partial-anonymity information design allows sellers to be more strategic than a social commerce marketplace with no anonymity. Further, with a partial-anonymity design, the likelihood of a successful transaction is higher than in a fully anonymous marketplace. Our results are robust to different product categories and uncertainty in product quality.

  • Strategic Best-Response Fairness Framework for Fair Machine Learning

    by Hajime Shimao, Warut Khern-Am-Nuai, Karthik Kannan, Maxime C. Cohen

    This study introduces a framework called “strategic best-response fairness” (SBR-fairness) to address discrimination perpetuated by machine-learning (ML) algorithms. It challenges the conventional focus on fairness solely in prediction results, arguing that this approach ignores how individuals affected by the predictions may alter their behavior in response to algorithmic decisions. The framework considers whether an algorithm, trained on potentially biased data, leads to identical equilibrium behaviors across different subpopulations that are ex ante identical. The study finds that common fair-ML algorithms, such as those relying on color-blindness and demographic parity fairness criteria, do not always achieve SBR fairness. This means that they may not eliminate disparities in effort and outcomes. Equalized odds (EO), however, have been shown to achieve SBR fairness, but they suffer from several practical limitations. The study proposes that SBR fairness is a necessary condition for breaking cycles of discrimination in ML. It also argues that SBR fairness offers a complementary way to assess other fairness criteria and understand behavioral responses. The findings suggest a need for policy and practical focus on designing SBR-fair algorithms that promote equitable outcomes at both the prediction and behavioral level.


  • The Impact of Online Q&As on Product Sales: The Case of Amazon Answer

    by Warut Khern-Am-Nuai, Hossein Ghasemkhani, Dandan Qiao, Karthik Kannan

    This study uses observational data from two online retail sites to examine the effect of questions and answers on sales of experience goods. Particularly, we leverage the naturally occurring experiment where the Q&A capabilities are available on only one platform. Interestingly, we discover that answers, especially the depth of answers, positively affect sales. Additionally, the fraction of questions with at least one answer positively and significantly affects product sales. Our additional textual content analyses unveil that fit- and quality-oriented questions have different effects on sales, as do positive and negative answers. Insights from this work can help platform managers develop Q&A ecosystems and content management policies.


  • Analyzing Frictions in Generalized Second-Price Auction Markets

    by Karthik Kannan, Vandith Pamuru, Yaroslav Rosokha

    Generalized second-price auction is the preferred mechanism for sponsored-search advertisements. In this paper, we take a twofold approach using Q-learning-based simulations and human-subject experiments to show that the low-value advertisers (who do not win the auction) exhibit highly exploratory behavior and impact the allocative efficiency of the mechanism. Moreover, we find the presence of bid-adjustment frictions (e.g., bid fee) moderates this phenomenon and results in higher allocative efficiency of the auction. Our focus on the bid-adjustment costs is motivated by the fact that both the sponsored-search platforms and policymakers can best observe and influence these types of frictions as compared with frictions that are difficult to observe (e.g., resources spent on the analysis of the market and sophistication of algorithms by the advertisers).


  • Top Cited Papers

    More than the Quantity: The Value of Editorial Reviews for a User-Generated Content Platform

    by Yipu Deng, Jinyang Zheng, Warut Khern-Am-Nuai, Karthik Kannan

    We investigate an editorial review program for which a review platform supplements user reviews with editorial ones written by professional writers. Specifically, we examine whether and how editorial reviews influence subsequent user reviews (reviews written by noneditor reviewers). A quasiexperiment conducted on a leading review platform in Asia, based on several econometric and natural language processing techniques, yields empirical evidence of an overall positive effect of editorial reviews on subsequent user reviews from the platform’s perspective. First, more reviews are provided for restaurants that receive editorial reviews. In addition, these reviews discuss substantive topics while also including a discussion on other topics, leading to a net increase in content length and variety. They also are more neutral in sentiment and are associated with lower rating valences. Further analysis of the mechanism reveals that the subsequent user reviews of the restaurants that receive editorial reviews become more similar to the editorial reviews in regard to topics, sentiment/rating, length, and readability, indicating a herding effect in how to write a review as the main driver of the change in the subsequent reviews. We further empirically isolate this herding effect among long-time reviewers. The findings suggest that review platforms could use an editorial review program not only to boost the quantitative aspect of user reviews but also, to manage the qualitative aspect as well. This paper was accepted by Kartik Hosanagar, information systems.


  • Identifying Perverse Incentives in Buyer Profiling on Online Trading Platforms

    by Karthik Kannan, Rajib L. Saha, Warut Khern-Am-Nuai

    Consumer profiling has become one of the most common practices on online trading platforms. Many platforms strive to obtain and implement technological innovations that allow them to understand and identify consumers’ needs, and, thereafter, monetize this capability by charging sellers to present and/or sell their products or services based on consumers’ interests. However, an interesting and relevant question arises in this context: Does the platform have an incentive to profile its buyers as accurately as possible? This paper develops and analyzes a parsimonious game-theoretic model to answer this research question. We find that, surprisingly, platforms that charge sellers for discoveries have a perverse incentive to deviate from accurate buyer profiling. However, such a perverse incentive does not exist for platforms that charge sellers for transactions. As a result, with such a perverse incentive, social welfare under discovery-based pricing is lower than that under transaction-based pricing.


  • A Graph-Based Ant Algorithm for the Winner Determination Problem in Combinatorial Auctions

    by Abhishek Ray, Mario Ventresca, Karthik Kannan

    Iterative combinatorial auctions are known to resolve bidder preference elicitation problems. However, winner determination is a known key bottleneck that has prevented widespread adoption of such auctions, and adding a time-bound to winner determination further complicates the mechanism. As a result, heuristic-based methods have enjoyed an increase in applicability. We add to the growing body of work in heuristic-based winner determination by proposing an ant colony metaheuristic–based anytime algorithm that produces optimal or near-optimal winner determination results within specified time. Our proposed algorithm resolves the speed versus accuracy problem and displays superior performance compared with 20 past state-of-the-art heuristics and two exact algorithms, for 94 open test auction instances that display a wide variety in bid-bundle composition. Furthermore, we contribute to the literature in two predominant ways: first, we represent the winner determination problem as one of finding the maximum weighted path on a directed cyclic graph; second, we improve upon existing ant colony heuristic–based exploration methods by implementing randomized pheromone updating and randomized graph pruning. Finally, to aid auction designers, we implement the anytime property of the algorithm, which allows auctioneers to stop the algorithm and return a valid solution to the winner determination problem even if it is interrupted before computation ends.


  • The Impact of an Augmented-Reality Game on Local Businesses: A Study of Pokémon Go on Restaurants

    by Vandith Pamuru, Warut Khern-Am-Nuai, Karthik Kannan

    Augmented reality (AR)–based applications have been growing in prominence. Pokémon Go, one of the most popular AR game applications, has been engaging in strategic partnership with businesses. The game combines geospatial elements with gamification practices to incentivize user movement in the physical world. In this study, we examine the value of such strategic partnerships between Pokémon Go and associated businesses. Specifically, we use data of the online reviews for a firm to study consumer engagement and consumer perception after the release of the game on its associated businesses. We find that restaurants associated with Pokémon Go do indeed enjoy a higher level of consumer engagement and more positive consumer perception. We also find that characteristics of the business significantly moderate this impact overall. Using a parsimonious analytical model, we also demonstrate the mechanism behind these results. Our study provides insights into consumer economic behavior in the context of AR applications. We further inform the business owners and policymakers regarding the potential value of such strategic business partnerships with such applications.


  • Welfare Implications in Intermediary Networks

    by Thành Nguyen, Karthik Kannan

    We study the welfare implications of competing middlemen in a two-sided market, where goods are intermediated between providers and purchasers. In our model, each intermediary sets the quantities it intermediates, and the prices are a consequence of a Cournot competition. Our analysis shows that, unlike traditional markets, increasing competition is not always beneficial for market efficiency and that mergers can have an ambiguous effect on efficiency. We also analyze how the underlying network influences social welfare. We define a parameter called the intermediary capacity of the network and show how the price of anarchy depends on this parameter. These results suggest an intuitive and simple measure for the level of competitiveness in a networked market involving intermediaries.