i-Cleantech Vlaanderen


Description


I-Cleantech Vlaanderen vzw is a cluster organisation in Flanders, the Northern part of Belgium. Together with companies, research institutions, public bodies and civil society actors, i‑Cleantech Flanders is a catalyst for innovation in a multitude of clean technologies and assists their subsequent implementation in society at large. More specifically, i‑Cleantech Flanders’ mission is “to identify and encourage the development of cleantech instruments that accelerate the realisation of a sustainable world”.

I‑Cleantech Flanders works cross-sectorally between existing organisations and focuses on four cleantech domains, being energy, water, materials and mobility. Central in i‑Cleantech Flanders’ structure are the pillars transition (management), research and industry.


Challenges


There is no simple data source available which defines clean technology, and which companies focus on cleantech, and how large a part of a company is actually involved in cleantech.

Even more in general there are no tools available to describe sub-systems in the economic tissue of a region, nor to describe the interaction within the economic ecosystem.

Thus, multiple, unstructured data sources must be combined into a knowledge base. Some of these sources are web pages or RSS feeds, which are analyzed using crawlers and semantic ontologies. Other sources are CSV files, relational databases, and web services.


Solution


The unstructured data is initially brought together in a relational database. This assumes the data can be fit into a predefined normalized model, which is not the case. Therefore, the future solution should be able to store unstructured data, and at the same time allow for ad-hoc querying and visualizations. It was decided to use a graph model for this, precisely because the focus of the project is to visualize relationships.

For the graph store, it was chosen to use the embedded Neo4J graph database. It is the most widely used graph database, and has an advanced and easy to understand query language, called Cypher. It also allows data visualizations through its internal web browser, and easy export of data for use in external graph visualization tools.