Read about OntoRefine - a new tool that allows you to do many ETL (extract, transform and load) tasks over tabular data.
Ontotext has recently released the latest 8.3 version of its signature semantic graph database GraphDB.
One of the key new features we’ve developed for this release is a wizard-like interface that guides users into creating various visualizations of RDF data with different starting points. You can configure the default graph visualization with the full expressivity of the SPARQL language to control what graph data you want to be displayed.
GraphDB now enables you to solve a lot of the complicated problems coming from dealing with real-world data. You have the power to control the starting point of the visualization as well as to create more than one visualizations over the same data. With this cool custom graph views, data exploration, data analytics and knowledge discovery become easier and faster. So you can use GraphDB’s power to infer relationships that are not explicitly stated in order to get the full picture of your data and obtain additional knowledge about the links in your datasets.
GraphDB is an RDF database, compliant with the RDF4J interfaces. One of its key features is that it is ready-to-use and that it lets you load tons of data from many different datasets. It also allows you to grow applications from the free version to the enterprise edition, which supports robust cluster and scaling.
Developing our semantic graph database is an ongoing process of refinement and we are always excited about the new features we are planning to add in the near future. We also offer excellent support to our community and commercial clients as well as a full suite of migration services to help you get started.
If you choose GraphDB as your semantic graph database, you can rely on a smooth experience throughout the whole cycle of working with data. GraphDB has the ability to take any structured data and generate an RDF out of it. Obviously, sometimes you would need to clean and transform the data first, which is covered by our OntoRefine interface, integrated into the workbench. Finally, you can expose your data as linked data.
So we have made the whole process easy for you – from starting with some data, doing data modeling and getting big data as a result.
At a later stage, you can also load any dataset from the Linked Open Data cloud or any other RDF datasets.
GraphDB 8.3 helps you configure the way you expand your visual graph. Based on our work with various customers, we have developed the most efficient way for you to control the visualizations. The powerful SPARQL language helps you model almost everything in your graph visualizations.
With the custom graph view configuration, you can choose to start with a search box, a fixed node or a SPARQL graph query result. Beginning with a search box means that you have one graph configuration, but each time you search for a different node to be your start node. Alternatively, beginning with a graph query result means that you can visualize everything that you can model with SPARQL as the initial state of your graph. This gives you a richer experience in exploring your data and finding hidden links.
There are also four different queries that allow you to configure the behavior and presentation of your graphs.
The first query specifies how new nodes and links are added to the visual graph when the user expands an existing node. The second one determines the node appearance – the text, color, size based on label, type and rank function. Next, you can control what data to appear for every node as metadata in the side panel. Finally, you can also choose how the predicates are labeled in these graphs.
With these four types of queries, you can work in many real-world data scenarios.
For example, sometimes you may want to integrate custom ontology schema that doesn’t follow the RDFS label or the SKOS schema. Here, GraphDB makes it easy by allowing you to choose which label to be displayed as the preferred label.
Alternatively, if you have provenance or metadata, you may want to remove it from your visualization. Or you may need to combine multiple links to generate new meaningful ones.
Finally, sometimes you want to filter some instances based on filters developed with SPARQL. Or, when not everything is properly modeled with the existing data, you can generate RDF resources on the fly.
The latest GraphDB release doesn’t only let you control the way you want to visualize your graph. Its new features also enable you to build more than one visualization on top of the same data and show you different types of nodes and relationships over it.
It’s also important to note that a visual graph configuration is more than a saved graph. A saved graph is just a snapshot of a graph. It is not functional as, without a graph configuration, you won’t be able to expand it.
Another cool thing is that GraphDB’s graph visualization has no limitations on the size of the data, as long as the database engine can query the information with SPARQL and there are no constraints in your web browser.
However, GraphDB’s visual graph has a maximum number of links to show because, even if your web browser allows it, it’s not sensible to have an infinite number of links. So, the more specific you make your queries, the more efficiently you can explore your data, based on what you are really interested in, rather than try and visualize as many links as possible.
GraphDB also allows you to visualize remote data. There are two ways to do that. The first one is to use the remote locations. You run a GraphDB server hosted somewhere else, and you can configure remote locations and connect with them. The other option is to use SPARQL that allows you not only to query remote data but also to do remote joins. Just keep in mind that, if you try to do remote joins, the performance of the SPARQL queries may be affected.
Now you see how GraphDB 8.3 and its new exciting visual graph creation capabilities puts you in control of what data you want to be queried, labeled and visualized. The power to create your own visual graphs enables you to explore all your datasets and most importantly, the links between your individual data nodes. GraphDB, coupled with SPARQL, gives you an all-round view of your information. And, as you know, the power of inference and custom visualization turns your information into knowledge and insights.
Want to create custom graph views over your RDF data?