D4.7 - LLM-based Retrieval Augmented Generation (RAG) System for Identifying Effective Data Monetisation Strategies

- Version Pending approval from EC commission - 

This deliverable builds upon the data valuation framework and decision-support tools previously developed within WP4 of the DATAMITE project, extending them through the integration of Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) techniques. The objective is to enhance the identification, recommendation, and contextualisation of effectivedata monetisation strategies by leveraging advanced AI-driven retrieval and reasoning capabilities. The developed RAG system operates over the DATAMITE tool ecosystem, enabling dynamic interaction between the data valuation taxonomy, KPI repository, and the Analytic Network Process (ANP)-based decision-support modules. By coupling domain-specific document retrieval with generative reasoning, the system can synthesise actionable insights, suggest optimal monetisation pathways, and explain metric interrelations in natural language. This approach allows stakeholders to query, explore, and refine monetisation strategies using intuitive conversational interfaces grounded in the validated knowledge base of WP4 outputs. Through this integration, Deliverable D4.7 demonstrates how LLM-based architectures can bridge human–machine understanding in data economy contexts, transforming static valuation frameworks into adaptive, intelligent assistants for strategic decision-making. The result is a cohesive and interoperable system that reinforces DATAMITE’s mission: empowering organisations to discover, quantify, and optimise the value of their data assets through trustworthy, explainable, and economically sound AI solutions.