We are excited to announce the release of metaphactory 5.5, which comes with a number of new features and enhancements across the entire spectrum of product capabilities, including improvements to metaphactory's semantic modeling capabilities, advancements to the platform's AI-driven features and the beta launch of a conversational interface, as well as enhancements to the model-driven app building functionality.

Enhancements to semantic knowledge modeling

Visual ontology editor

• Simplified customization and maintenance of the ontology editor, allowing knowledge graph engineers to configure custom metadata properties on an ontology or on any ontology element (for example, to support synonyms, scope notes, etc). This also results in a significant reduction in the time needed to update to new product versions and results in a simplified maintenance of customizations.
• Ability to batch create ontology elements (classes, relations, attributes), significantly streamlining the modeling task for knowledge stewards.
• Improved ontology editor user experience (UX), providing users with a simplified interaction and a more coherent editing experience for ontology metadata.
• Simplified ontology import/export workflows, allowing knowledge stewards and knowledge graph engineers to import and export ontologies in a single action, as well as import/export ontology diagrams alongside the ontology itself. This reduces the effort required to transport ontologies and corresponding diagrams across metaphactory instances.

Quality assurance for ontologies and vocabularies

• Ability to monitor ontology and vocabulary quality, enabling the management and review of assets that may have been created or modified outside of metaphactory. This further supports knowledge stewards in adhering to common best practices and ensuring that the onologies they create are valid.
  
  Additionally, knowledge graph engineers benefit from more complex and integrative data quality checks that complement the existing functionality in the visual editor, as well as new functionality providing differentiation details with regards to severity. This supports governing best practices, considering the diversity of actors in the enterprise.
  
  The data quality checks are implemented in standard SHACL rules. metaphactory 5.5 provides a default set of rules following community best practices, which can be easily extended to customer or vertical-specific needs.

Integration of modeling events into enterprise systems

• Ability to subscribe to modeling event notifications, allowing to send users or downstream systems updates on any actions or status changes related to the workflow or lifecycle of an ontology or a vocabulary. This is powered by a new event bus.
• A sample app to demonstrate the integration in Amazon Simple Notification Service (SNS). The integration in Amazon SNS allows metaphactory to send notifications, e.g., via email, for example, if a new review is requested for a given ontology.

LLM-based Interface to the Knowledge Graph

• Enhancements to the LLM-based natural language to SPARQL (NL2SPARQL) translation component, further supporting end users who want to ask natural language questions over the knowledge graph by delivering better translation quality as a result of the improved entity disambiguation based on type detection. Additionally, these enhancements simplify the process for application engineers when configuring the component to particular use cases and verticals. This functionality is now available out-of-the-box with metaphactory, without the need for any additional container and deployment steps.
• Conversational Interface utilizing LLMs - Beta version, allowing end users to lead contextualised conversations with the graph. The conversational interface is enabled by a range of tools and services that allow answering queries of different kinds over the knowledge graph, including the generation, refinement and execution of structured queries as well as answering unstructured queries using a RAG-approach (Retrieval Augmented Generation). Depending on the user question, the conversational interface selects and orchestrates the adequate services to answer the question and render the results in natural language or structured format for further processing.

metaphactory Conversational Interface - Beta version

Miscellaneous

• New experimental mechanism to support application-specific annotations to ontologies, further enhancing metaphactory's capability to generate application logic from models. Through these annotations, the model can be used to generate user interfaces that are more tailored to specific application or user needs, while minimizing the need for creation and maintenance of custom code.
• Model-driven, transparent federation
• New query planning and execution optimizations contributing to the overall performance and stability of metaphactory's transparent federation capabilities
• Ability to the model-driven source selection to read relevant information dynamically from the knowledge graph using a SPARQL query
• Support for aliasing of federation members
• First version of metaphacts semantic modeling guidelines, enabling users who are new to the modeling world to learn the basic concepts involved and get started with modeling on their own. The guide introduces the concept of modeling and the basic elements in semantic modeling language, and discusses initial methodology considerations and design principles for ontology modeling.
• Best practices to work with lexical variations in keyword search and entity lookup

Have a look at the release notes or This email address is being protected from spambots. You need JavaScript enabled to view it. to learn more about this release.

Get started with metaphactory for free »