Quality and Usability

Privacy, Security and Trust in Crowdsourcing Confidential Enterprise Data


Aim of this project is to evaluate and enhance the usage of crowdsourcing micro-task services by enterprises considering data confidentiality and providing tools and solutions for that.

The focus is on 1) which tasks can/cannot be assigned to a specific worker and 2) who (workers) should be permittd to perform a task, both with a goal of increasing data confidentiality. The confidentiality is algorithmically improved by dispersion (maximize unrecognizability) using spatial assignment and disruption functions. Individual software solutions for confidential crowdsourcing are  developed and evaluated for spatial work assignment and user scoring/ selection.

Business cases are evaluated and results should be transfer to Web-based Starbytes (REPLY) and Mobile-based Crowdee (TUB). Partners also working on spoof-prevention of location information using continues (multiple) measurements, from multiple sources (e.g. GPS, IP, WiFi-prints) of user position. The goal is to detect fake locations.

Meanwhile worker solutions for inviting specific groups of workers to perform tasks are developed and evaluated. Workers are selected based on their social and behavioral scores assigned to them by interaction modeling and gamification depending to the work-type.

Expected Results:

  • Privacy-aware crowdsourcing solutions including spatial and selection confidentialiy strategies
  • Security-aware crowdsourcing and spatial spoofing prevention
  • Integration, business modeling and going-to market for proposed solution in mobile- and web-based crowdsourcing environments

Duration: 01/2016 - 12/2016

QULab  members: Tim Polzehl, Babak Naderi

Partner: ELTE - Eötvös Loránd University, Hungary

Industrial partner:Starbytes, REPLY, Italy

Funded by:EIT Digital