Making Data More Accessible and Usable: Knowledge Graphs, Semantic Layers, and Vector Databases

CASE STUDY

A global foundation had a mandate from its CEO for data teams to “figure out a way to adopt AI.” Specifically, the mandate was to use LLMs to evaluate the impact of its investments on strategic funds by synthesizing information from publicly available domain data, internal project reports, dashboards, and internal investment/financial data. The initial challenge for previously failed efforts lay in successfully linking diverse, unstructured information to structured data and ensuring the insights generated were explainable, reliable, and actionable for executive stakeholders.

To address these challenges, we leveraged a vector database that was augmented through advanced graph technology and a RAG workflow backed by a semantic layer. To provide the relevant organizational metrics and connection points in a structured manner, the solution leveraged an investment ontology as a semantic backbone that explicitly defines how the data entities within the disconnected investment data are related, from structured datasets to narrative reports, and harmonized under a common language. This semantic backbone supports both precise data integration and flexible query interpretation.

To effectively convey the value of this hybrid approach, we leveraged a chatbot that served as a user interface to toggle back and forth between the basic vector model and the GraphRAG solution. The result was a solution that consistently outperformed the basic/naive vector RAG for complex questions, demonstrating the value of semantics for providing organizational context and alignment. Ultimately, it delivered coherent and explainable insights that used aligned terms to bridge structured and unstructured investment data and provided a transparent AI mapping that allowed stakeholders to see precisely how their AI solution derived each answer.

This case study is just one example of why organizations are now seeking and investing in a coherent, integrated way to understand their vast data ecosystems. Because organizations often work with complex systems, ranging from CRMs and ERPs to data lakes and cloud platforms, extracting meaningful insights from this data requires a coherent, integrated view that can bridge these gaps. Knowledge graphs provide a map of relationships, semantic layers bring depth of understanding, and vector databases unlock speed and scale in information retrieval. Individually, each of these technologies solves a piece of the data accessibility puzzle.

When integrated, they provide a pragmatic approach that enables organizations to bridge these gaps, streamline processes, and transform how data is used across teams, departments, and bespoke systems.