A Privacy-Preserving and Efficient Spatial Keyword Based Task Matching Scheme in Crowdsourcing

Crowdsourcing has emerged as a vital paradigm for task execution and data collection, with task matching as a core application. crowdsourcing platforms can leverage the spatial keyword similarity to identify whether a worker’s interests and location align with a task requester’s requirements. However, untrusted crowdsourcing platforms pose significant privacy risks to both task requesters and workers. To mitigate these risks, participants typically encrypt their data before outsourcing. In this paper, we propose a privacy-preserving Spatial Keyword Similarity-based Task Matching (SKSTM) scheme that enables secure task matching. In SKSTM, we encode both locations and keywords using Geohash and bitmap representations, respectively, transforming secure task matching into inner product operations in the ciphertext domain via Enhanced Asymmetric Scalar Product-Preserving Encryption (EASPE). Security analysis and experimental results demonstrate that SKSTM preserves participants’ privacy while outperforming state-of-the-art schemes in task matching efficiency.