The main objective of the LUCERNA project (Luce IT) is to research and design a technological framework for data availability, service optimization and big data analytics based on semantic data lakes, deep neuroevolutionary algorithms and explainable artificial intelligence. This new methodological and technical framework will serve to demonstrate that it is possible to design intelligent solutions for the capture, storage, processing and analysis of large volumes of data in an agile way and independent of the original business model. The main result of this research project will be, therefore, a framework based on a set of virtual organizations of light agents oriented to the ingestion, securization, processing and analysis of large volumes of data that will be previously characterized by means of ontologies and semantic annotation techniques in order to abstract the data model from the business model of the different user entities in each problem to be addressed. Likewise, the framework will allow the transition from massive unstructured data in the form of semantic data lakes to structured repositories and will provide the user entity with different services and resources associated to its data by means of neuroevolutionary algorithms that optimize their availability. By means of knowledge graphs, convolutional networks, grouping techniques based on spectral clustering and explainable artificial intelligence techniques, each entity will be able to extract the most relevant characteristics and obtain knowledge about its data, as well as generate explanations that allow understanding the models and visualizations obtained using natural language. This new framework will be validated in the LUCERNA project (Luce IT) by means of a demonstrator in a laboratory environment, thus reaching a TRL4 maturity level.
