PHASE 1: Concepts

  • This phase will study the general problem, the scope of the project (which data analytics routines we will focus on), use cases, the actors, the trust/adversary model, and the requirements (user and technical requirements).
  • This phase will also review the state of the art in privacy preserving data analytics and identify the limitations of existing techniques with respect to the requirements, the use cases and the adversary model.

PHASE 2: Design of the modules and the platform

  • We will first design the “atomic” privacy preserving data analytics modules. Based on the identification of the gaps existing between previous work and the requirements, we will select or develop new cryptographic modules that solve the problems characterized in PHASE 1.
  • In parallel, this phase will delineate the architecture of the integrated platform that will combine the several modules in a single offering.
  • Moreover, we will specify the design components of the dashboard of the platform with respect to the requirements and the use cases defined in PHASE 1.

PHASE 3: Implementation of the modules and the platform

  • We will deploy the modules, the underlying cryptographic primitives, and the platform. We will make the suitable implementation choices (programming language, framework, software, hardware, or any other available tools, etc.) to deliver solutions that practically satisfy the requirements identified in PHASE 1 and which adopt the design approaches characterized in PHASE 2.

PHASE 4: Validation against the use cases

  • A proper validation roadmap will be first defined. We will also identify the validation criteria.
  • The actual validation procedure will be performed. Conclusion and recommendations will be derived.

PHASE 5 (transversal PHASE): Dissemination and exploitation.

  • Throughout the project, we will promote the project’s objectives, innovation aspects and results in order to ensure awareness of the target communities with respect to privacy preserving data analytics.
  • PAPAYA results will be promoted within industrial partners (MCI, ORA, IBM, and ATOS) to ensure the alignments between industrial requirements and the project's outcomes so as to ease integration.
  • We will also aim at developing and maintaining a sustainable exploitation plan of the project’s outcomes and at initiating knowledge transfer towards the target stakeholders.