Modeling and mitigating human annotation errors to design efficient stream processing systems with human-in-the-loop machine learning

•Study of human annotation task in hybrid stream processing systems.•Presenting a generic human error framework of serial ordering-based mistakes and slips.•Verifying of the proposed human error framework through extensive experiments.•Presenting a novel method for human error-mitigation in an active learning paradigm.•Validating the novel method through simulation-based experiments.