Graphwise AI Summit

A true semantic layer suggests a different development paradigm, a desiloed and thus rearchitected paradigm that dovetails nicely with agent-based development. Engineers and developers will need the semantic layer’s monitoring, articulation and control capabilities to manage the agent environment, in fact.

This talk will cover ten primary characteristics that define a true semantic layer. It will also review other claimed semantic layers that by comparison lack sufficient desiloing, power and cohesion, and will explore why they don’t measure up. These will not have the staying power and vision that 20+ years of semantic standards have already demonstrated.

Regulatory compliance, particularly in the rapidly evolving domain of ESG, has become a formidable challenge. The sheer volume of documents, coupled with the complexity of legal frameworks, makes manual analysis a cumbersome and inefficient process.

This presentation will introduce an innovative approach to this problem that leverages the power of AI and knowledge graphs. We have developed a tool that transforms unstructured documents into a precise, structured knowledge graph. By using advanced AI techniques to extract and contextualize information, our solution provides a quantifiable understanding of what each report discusses and how it aligns with relevant frameworks.

We will illustrate this methodology through the practical use case of analyzing and benchmarking ESG reports. Based on a precisely modeled ESG taxonomy, the application makes it easy to analyze what a report talks about, where the relevant paragraphs are, how it relates to regulatory frameworks and how it compares to industry peers reports. Additionally, GraphRAG applications make use of the data to give curated insights based on selected topics and reports.

You will see how this solution has been implemented using the Graphwise Platform to turn a compliance burden into a strategic asset.

Every organisation today seems to be talking about data products – but without a shared language, the term risks becoming hollow. DPROD, the Data Product Ontology, was created to change that: it provides a standards-based way to describe, govern, and connect data products across domains.

In this talk, I’ll introduce the idea of Semantic Data Products – data products built on DPROD that carry meaning as part of their design. By embedding semantics, they bring clarity, consistency, and interoperability to fragmented data landscapes. Most importantly, they invert the cost of integration: instead of consumers bearing the burden, producers publish self-describing, linkable, and AI-ready data products from the start.

Is much of your valuable enterprise knowledge stored in SharePoint?

Discover how AI is transforming workflows to improve content discoverability, boost agent reliability, and manage accuracy at scale.

This session will showcase how Altuent is using Graphwise for M365 and Copilot agents to build intelligent taxonomies, automate tagging, and detect ambiguity.

We’ll also demonstrate how an AI-enabled workflow can help manage and scale content accuracy efforts, empowering Copilot agents to surface the reliable information users need.

Beamery is a London-based scaleup operating in the HR domain, delivering cutting-edge AI solutions for talent management. This session will provide an insight into why GraphDB was selected as the application database for Beamery’s flagship product.

Specifically:

• The benefits of using a graph in the HR space
• Enterprise-grade functionality of GraphDB
• GraphDB as a GraphQL API server for multiple products
• Our experiences operating GraphDB in a multi-tenant environment

The talk focuses on the challenges of producing and consuming product data in the current business and regulatory landscape and how semantic web technologies (knowledge graphs and taxonomies) can help to overcome these challenges ensuring semantic interoperability.

It also provides insights on how Digital Product Passports from different sectors can be interoperable and constitute an added value for the whole market.

In-vivo preclinical animal studies are vital for drug discovery, but fragmented data solutions hinder exchange and reuse, creating challenges for data scientists. The FISH (FAIR In-vivo data SHaring) platform is a “FAIR by design” system that unifies siloed data into a single platform with standardized, context-rich metadata. Utilizing globally unique persistent resolvable identifiers (GUPRIs), FISH ensures seamless data access and exchange.

The platform employs semantic models and standardized terminologies for structured, consistent, and machine-actionable data capture, integrating with existing LIMSes to prevent data duplication. FISH leverages Graphwise’s GraphDB and Semantic Objects solutions to provide a FAIR GraphQL API, exposing data from the metadata store to data scientists and downstream systems with access control. By FAIRifying data, FISH maximizes the potential of animal studies, enabling data reuse, efficient ML/AI applications, and workflow automation. This approach enhances reproducibility, facilitates data repurposing, increases the probability of successful Entry into Human (EiH) trials, and reduces the need for additional animal testing.

The pharmaceutical industry is at the forefront of developing precision medicines. Gone are the days when we broadly classified disease, now we have the understanding to characterise diseases to a fine grain. For example we once talked about cancer, then cancer of an organ, then cancer of a cell type and now we define cancers by specific mutational profiles. As our scientific understanding has grown we need to re-exam our previously collected clinical data to identify these subpopulations and create virtual cohorts of multi-modal (clinical observations, medical images, digital pathology, DNA sequencing, protein expression profiles, etc) from across multiple clinical trials that can be used to further understand a disease and enable the development of improved treatments for patients.

We have solved this challenge by delivering a semantic knowledge graph that integrates multiple data modalities from a range of diverse sources together. This information is then exposed via a user interface. Scientist are now able to perform tasks in minutes that previously took weeks or months of manual work.

This presentation will describe the challenges overcoming inconsistencies across data sources (>20 sources), dealing with massive data volumes (over 10 trillion triples), creating intuitive interfaces for a large complex graph and why we still need librarians.

Notarial deeds are a cornerstone of legal certainty in the Netherlands, rich in meaning, but traditionally crafted for human interpretation. As part of its digital transformation, the Royal Dutch Notarial Association (KNB) is working with Graphwise to extract structured legal knowledge from these documents using graph technology and AI.

In this session, KNB Architect Wouter Siepman will share how the organization is developing semantic representations of notarial texts. The goal is to identify key entities such as persons, legal objects, and juridical acts, and map their relationships within and across transactions. This approach not only unlocks historical data, but also supports explainable and trustworthy AI applications in the legal domain.

The talk will also highlight how this effort fits into KNB’s broader architectural vision: building a central knowledge structure for the notarial ecosystem that formalizes and preserves deep, often tacit, domain knowledge. This work enables innovation while maintaining compliance, integrity, and professional standards.

For anyone exploring the application of knowledge graphs, entity extraction, or AI in regulated, high-trust environments. This session offers concrete insights from a real-world legal transformation journey.

The requirements for the provision of content for technical communication have steadily increased in recent years. Manufacturers are becoming service companies and documentation is gaining a relevance that goes beyond the fulfilment of legal obligations. The shortage of skilled labour means that technical documentation is also becoming relevant in the context of training and further education activities. This means that, on the one hand, the user-related or context-related provision of content is becoming increasingly relevant and, on the other hand, the combination of content from different sources is becoming more and more important.

We will show use cases out of a PoC from Sandvik using poolparty on how the standard iiRDS can be used as the backbone for delivery of intelligent content (as the basis for digital twins) that offer content as a service by combining content, knowledge graphs and LLMs.

Over the past two years, Graphwise has partnered with EY on their Discover Reimagined project, a key initiative aimed at revolutionizing the firm’s global knowledge ecosystem. By integrating advanced technologies like knowledge graphs and large language models (LLMs), the project is designed to give EY’s client-facing teams seamless access to critical information, bridging gaps across service lines and streamlining knowledge management workflows.

The goal? To ensure that the right content reaches the right people in the right context—empowering smarter decision-making and delivering better outcomes for EY’s clients.

In this presentation, we’ll take you through our journey of building and evolving this cutting-edge platform. We’ll explore the challenges we’ve overcome, share the lessons learned, and discuss the measurable impact it’s had on the business.

But the conversation doesn’t end there. The future of knowledge hubs is just beginning, and we want you to be part of it. Join us for an interactive design thinking session, where we’ll collaborate and brainstorm ideas on what the knowledge hub of tomorrow could look like. Together, let’s shape the future of knowledge sharing in an AI-driven world.