With the rapid development of blockchain technology, the data element transaction market is facing unprecedented opportunities and challenges. This study proposes an innovative data element transaction model that integrates zero-knowledge proof technology on the blockchain with game theory, aiming to address information asymmetry effectively in data transactions. In traditional data trading scenarios, sellers typically possess more information about the data source, quality, legality, and potential value, whereas buyers find it challenging to assess this information accurately. This not only increases the decision-making difficulty and transaction risk for buyers but also may lead to the “lemons market” phenomenon—where high-quality data gradually exits the market owing to the inability to obtain a price that reflects its value, whereas low-quality data dominate, ultimately harming the overall reputation and efficiency of the market. This model uses zero-knowledge proof technology to analyze the behavior patterns of transaction parties through game theory methods without revealing any additional information, thereby optimizing transaction strategies and increasing the fairness and success rates of transactions.