Franz Releases the First Neuro-Symbolic AI Platform Merging Knowledge Graphs, Generative AI, and Vector Storage
Large language models (LLMs) have gained significant popularity due to their vast potential. However, they also come with limitations such as a lack of explainability and the inability to validate information from external sources. To tackle these challenges, Franz, a leading supplier of Graph Database technology and an early adopter of AI, has unveiled AllegroGraph 8.0. The latest version integrates new Knowledge Graph capabilities to enhance accuracy, transparency, and explainability in LLMs.
The pioneering Neural-Symbolic AI platform incorporates LLM components directly in the SPARQL, integrating vector generation and vector storage to create an extensive AI Knowledge Graph. By combining machine learning with knowledge and reasoning capabilities, AutoGraph 8.0 can now tackle complex problems. Furthermore, the platform’s ability to generate a Knowledge Graph with lesser data has expanded its applications to various use cases.
Already trusted and used by several Fortune 500 companies globally, AllegroGraph 8.0 aims to strengthen Franz’s position in the industry, following the recognition gained with AllegroGraph 7.0.
Dr. Jans Aasman, CEO of Franz Inc., highlighted the challenges faced by LLMs, stating, While general-purpose LLMs excel at straightforward tasks that do not necessitate background or changing knowledge, addressing more complex, knowledge-intensive queries demands the capabilities provided with a Knowledge Graph to avoid generating ‘hallucinations’. Hallucinations refer to data that appears real but lacks grounding in factual information. This issue can lead to contextual ambiguity, bias amplification, and an inability to verify data accuracy.
To combat these concerns, AllegroGraph 8.0 introduces Retrieval Augmented Generation (RAG) capabilities, combining seamless Generative AI within a Knowledge Graph platform. This integration ensures that LLM outputs are fact-checked and grounded in knowledge-based information.
One of the standout features of AllegroGraph 8.0 is the new LLMagic function, enabling users to engage with the platform through dynamic natural language interfaces for both reasoning processes and queries. This conversational approach mirrors human conversation, enhancing productivity for existing users and expanding accessibility for new users.
The platform also introduces VectorStore tools, bridging the gap between Knowledge Graphs and enterprise documents. Users can now access knowledge contained within documents, gaining profound insights and a comprehensive understanding of enterprise data.
For AI symbolic reasoning, the upgraded platform includes a built-in rule-based system that users can configure. This empowers users to transform complex data into interpretable and actionable rules. Symbolic rule generation enables prediction and classification based on data, providing transparent explanations expressed through symbolic rules. This feature enhances interpretability and trust in AI systems.
AllegroGraph 8.0’s LLMagic tools streamline time-consuming and intricate tasks involved in creating taxonomies and ontologies. By analyzing diverse data and identifying relationships, patterns, and semantic connections, the platform generates structured hierarchies and classifications.
Additionally, the upgraded AllegroGraph 8.0 offers enhanced scalability and performance, a new web interface, and advanced Knowledge Graph visualization through integration with Franz’s renowned graph visualization software, Gruff v9.
As the demand for LLMs continues to grow, the release of Franz’s AllegroGraph 8.0 signifies a significant step towards addressing the limitations associated with these models. By merging Knowledge Graphs, Generative AI, and Vector Storage, Franz aims to provide a transparent, accurate, and explainable AI platform. This advancement promises to revolutionize various industries and empower organizations to leverage the immense potential of large language models.
