Ontotext is pleased to announce the release of GraphDB 10.3.
GraphDB 10.3 introduces several exciting new features. One notable addition is the integration of ChatGPT, a popular Large Language Model, which greatly assists users in comprehending SPARQL queries and results, while also providing a new mechanism to enrich their RDF data.
The upgrade of Ontop, a powerful solution for data virtualization, to version 5 significantly reduces the time and effort required to construct a knowledge graph from existing data sources by bringing new features and support for additional data sources. Upgraded versions of the Elasticsearch and Solr connectors, along with a new dedicated OpenSearch connector, offer users better flexibility for indexing and searching their data. GraphDB 10.3 also includes numerous other improvements and bug fixes, further enhancing the overall user experience and stability of the system.
GraphDB 10.3 features an innovative integration with ChatGPT, one of the most popular Large Language Models. This new feature is designed to revolutionize the way users interact with RDF and unlock its full potential for both newcomers and experienced users.
With the integration of ChatGPT, users will now be able to delve into the world of RDF with greater ease and gain a deeper understanding of its capabilities. Whether they are just starting out or have advanced knowledge, this integration empowers them to leverage ChatGPT’s capabilities and seamlessly integrate them with their RDF data to extract valuable insights.
One of the most significant aspects of this integration is the ability to explain queries and/or results with the assistance of ChatGPT. This feature ensures that users not only obtain accurate and relevant answers but also gain a better understanding of the underlying query logic and the significance of the obtained results. Such explanations foster a more informed decision-making process and empower users to extract maximum value from their RDF datasets.
Furthermore, GraphDB now provides the ability to pose generic questions to ChatGPT directly from SPARQL. By doing so, users can enrich their graphs and extract additional information from their data. For instance, they can enrich the descriptions of companies in their graph with information about their industries or products. This novel approach brings a new level of interactivity and intelligence to RDF queries, opening up endless possibilities for knowledge discovery.
Data projects often involve existing information sources that employ non-semantic data models or are too large to be replicated, making the extract, transform, load (ETL) approach infeasible. This can pose challenges for users and limit their confidence in building a robust knowledge graph from these sources.
To address this, Ontotext has made substantial advancements in integrating GraphDB with various external data sources, enabling seamless connectivity and simplifying knowledge graph construction. In particular, the upgrade to Ontop 5 brings expanded integrations with well-known platforms such as Dremio, Databricks, and Snowflake. These integrations have been carefully designed to streamline the process of incorporating data from these platforms into their knowledge graph via data virtualization or partial replication.
By harnessing the power of GraphDB’s upgraded Ontop integration, users can now enjoy reduced time and effort when building their knowledge graphs. This means they can focus more on extracting insights and deriving value from their data, rather than struggling with complex data integration challenges.
Many users rely heavily on externally managed full-text search (FTS) services such as Elasticsearch and Solr. In response to the growing demand and the need to upgrade to the latest major versions, Ontotext has upgraded the Elasticsearch and Solr connectors in GraphDB to the latest client versions. This ensures seamless compatibility and empowers them to leverage the advanced features and improvements offered by the latest major releases.
Also, in recognition of the rising popularity of OpenSearch as an alternative to Elasticsearch and since the two platforms have diverged substantially, Ontotext has taken a proactive approach to provide a dedicated implementation for OpenSearch. This dedicated connector implementation enables seamless integration of OpenSearch into their GraphDB workflow, giving greater flexibility and options for indexing and searching their data, and providing access to the same features as the Elasticsearch connector.
As a general strategy to offer a secure and reliable product, Ontotext strives to provide up-to-date versions of third-party libraries. This includes both features and bug fixes provided by the libraries and also addresses newly identified public vulnerabilities.
The RDF4J library in GraphDB was upgraded to 4.3.3 and brings a number of SHACL improvements and general bug fixes.
For more information, contact Doug Kimball, Chief Marketing Officer at Ontotext