Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

SEPAR: A Privacy-Preserving Blockchain-based System for Regulating Multi-Platform Crowdworking Environments

Abstract : Despite recent intensive research, existing crowdworking systems do not adequately address all the requirements of a real-world crowdworking environment. First, crowdworking platforms need to integrate within society and in particular to interface with legal and social institutions. Global regulations must be enforced, such as minimal and maximal work hours that participants can spend on crowdworking platforms. Second, crowdworking platforms are naturally distributed and need to collaborate with each other to process complex tasks, resulting in the rise of multi-platform crowdworking systems. Moreover, while collaborating to enforce global regulations or while processing complex tasks that require the transparent sharing of information about the tasks, the system needs to preserve the privacy of all participants. In this paper, we present SEPAR, a multi-platform crowdworking system that enforces global constraints on distributed independent entities. In SEPAR, Privacy is ensured using lightweight and anonymous tokens, while transparency is achieved using a permissioned blockchain shared across multiple platforms. To support fault tolerance and support collaboration among platforms, SEPAR provides a suite of distributed consensus protocols. The privacy guarantees of SEPAR against covert adversaries are formalized and thoroughly demonstrated, and the experiments reveal the efficiency of SEPAR in terms of performance and scalability.
Complete list of metadatas
Contributor : Tristan Allard <>
Submitted on : Monday, May 11, 2020 - 6:00:42 PM
Last modification on : Saturday, July 11, 2020 - 3:16:14 AM

Links full text


  • HAL Id : hal-02570108, version 1
  • ARXIV : 2005.01038


Mohammad Javad Amiri, Joris Duguépéroux, Tristan Allard, Divyakant Agrawal, Amr El Abbadi. SEPAR: A Privacy-Preserving Blockchain-based System for Regulating Multi-Platform Crowdworking Environments. 2020. ⟨hal-02570108⟩



Record views