Workpackages

INODE is structured in 10 Workpackages, built around the envisaged system architecture.

WP1 provides project management; WP2 defines the overall system architecture; WP3 provides the core services for federated query processing; WP4 enables the vision of OpenDataLink; WP5, WP6 and WP7 enable the vision of OpenDataDialog; WP8 will evaluate the services of OpenDataLink and OpenDataDilog; WP9 integrates and tests software components; WP10 is responsible for dissemination and training.

To ensure seamless integration across WPs, we will use a SCRUM-inspired agile software development approach throughout the whole project, by maintaining a code repository and automating the build, testing and deployment of the services.

Work Package #1

This work package provides professional project management to all partners and fosters efficient communication. It oversees the various project phases, tracks the progress of the project and provides regular progress reports to the European Commission in a timely manner. Moreover, WP1 continuously monitors, tracks and mitigates potential issues/risks and makes sure the budget constraints are fully met.

PARTICIPANTS:

ZHAW (Lead), ATHENA, MPI, FHG, SIRIS, CNRS, UNIBZ, SIB, Infili

Work Package #2

This work package establishes the major project requirements and defines the use cases in close collaboration with the industry partners as well as the implementation of work packages (WP3-WP7). Three use cases will serve as a yardstick for measuring the functionality as well as the success of the project. The use cases cover the following application domains: bioinformatics, astrophysics as well as research and innovation policy making. Based on the requirements, we will design the best suitable architecture.

PARTICIPANTS:

ZHAW, ATHENA, MPI, FHG, SIRIS, CNRS,UNIBZ, SIB (Lead), Infili

Work Package #3

This WP designs and implements the technology that INODE relies to provide an integrated coherent view of heterogeneous data sources, and to enable OBDA and efficient processing of rich analytic queries generated by the data access and exploration services.

PARTICIPANTS:

ZHAW, ATHENA, FHG, SIRIS, UNIBZ (Lead), Infili

Work Package #4

In this WP, we will develop the services that enable linking of loosely coupled collections of data sets to support queries across them. For mapping construction, we will follow two approaches, namely data-driven and task-driven mappings, and develop supporting technology. Moreover, we will develop techniques for automatically constructing a knowledge graph out of text data, using a machine learning approach that identifies entities and relationships in unstructured documents.

PARTICIPANTS:

ZHAW, ATHENA, SIRIS, UNIBZ, Infili (Lead)

Work Package #5

This work package implements by-example, by-analytics and by-natural language explorations. In by-natural language, the user provides a query in natural language and INODE translates the query into SQL or SPARQL – depending on the underlying data sources.

PARTICIPANTS:

ZHAW (Lead), ATHENA, FHG, CNRS

Work Package #6

This WP designs and implements the technology for building a conversational experience between the user and the system. It will implement data exploration insights and NL explanations.

PARTICIPANTS:

ZHAW, ATHENA (Lead), FHG, CNRS, Infili

Work Package #7

This work package provides services to explore search results and enable interactive query manipulation that go beyond lists of results, and beyond text input fields for querying. We are looking for visualizations that put queries, search results and user session history in context, emphasizing relations between items of interest, so that the user yields clues for proceeding with the exploratory search and deciding whether the exploration has finished.

PARTICIPANTS:

ATHENA, MPI, FHG (Lead), SIRIS, SIB, Infili

Work Package #8

We will perform a detailed evaluation of INODE using quantitative and qualitative approaches. As it is easier to start with a quantitative evaluation, we will define for each use case, a set of measures. The qualitative and the user experience evaluations are on the other hand dependent on users and will require to design hand-crafted studies and to recruit relevant users. This work package will be developed in conjunction with WP5 and WP7 to ensure consistency in the measures and approaches used throughout the project.

PARTICIPANTS:

ZHAW, ATHENA, MPI, FHG, SIRIS, CNRS (Lead), UNIBBZ, SIB, Infili

Work Package #9

This work package integrates, tests and evaluates the software components delivered by the implementation packages WP3 to WP8. We have opted for agile development with several software development sprints. The assumption is that the components delivered by WP3 to WP8 have already passed the unit tests. WP9 is responsible for functional testing, integration testing, performance testing and acceptance testing in close collaboration with the implementation work packages and the use case providers. In addition, WP9 is also responsible for integrating all delivered software components with the EOSC-hub.

PARTICIPANTS:

ZHAW, ATHENA, MPI, FHG, SIRIS(Lead), CNRS, UNIBBZ, SIB, Infili

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 863410