Adaptive Pricing in Crowdsourcing Platforms Using Prospect Theory and Extensive-form Game Models

Crowdsourcing platforms, as new labor markets, have been widely attracting the attention of businesses and researchers because they provide more efficiency. Despite this popularity, they do not readily bring about success as their functionality and efficiency in attracting users are influenced by factors such as users’ willingness to participate and user preferences. Previous studies have investigated how monetary rewards can affect the users’ willingness to participate in crowdsourcing platforms. However, the role of user preferences and features on the design of pricing mechanisms and, in turn, users’ willingness to participate is understudied. This study draws upon prospect and extensive-form game theories— which can model the interaction between users—to propose and test an adaptive pricing method wherein understudied factors such as skill level, reputation score, required activity, and monetary reward are considered to provide optimal pricing solutions so that both the requesters and crowdworkers are satisfied more. The proposed method was tested with real interactional data from a micro-tasking translation platform in Iran. The results showed that the proposed algorithm cannot only increase participation but also enhance the platforms’ profit. This study provides valuable guidance for platform managers and designers to remain adaptive, ensuring their strategies match with evolving users’ behaviors.