Handling uncertainty in citizen science data: Towards an improved amateur-based large-scale classification.

•Citizen Science is a promising data source for many real-world applications.•Citizen Science outcomes show a generalised lack of accuracy and validation.•A novel multi-stage approach to handle uncertainty in amateur-based classification.•A case study with Galaxy Zoo project data is used to test the proposed method.•Experiments show an improvement in accuracy and number of examples classified.