Learned-Index-Based Semantic Keyword Query on Blockchain

Blockchain has become increasingly popular for data management in recent years. However, the existing blockchain systems lack efficient semantic queries, particularly keyword queries. To address this issue, we propose a learned-index-based query architecture on blockchain. First, our records semantics information to support Second, establish lookup table index among blocks and block-level recursive model improve efficiency. We store extended block headers maintain result’s completeness, indexes off chain optimize maintenance Third, verifiable algorithm based proposed correctness. Finally, experimental results show that combining learned effectively improves efficiency