Semantic Technologies for supporting KDD Processes
Achieving a comfortable thermal situation with an efficient use of energy remains still an open challenge for most buildings. In this regard, the advent of the IoT (Internet of Things) and maturity of KDD (Knowledge Discovery in Databases) processes may contribute to the solution of these problems. However, the adequate combination of these two technologies is not straightforward, due to the heterogeneity and volume of the data to be considered. Therefore, data analysts could benefit from an application assistant that supports them throughout the KDD process.
This research work aims at supporting data analysts through the different KDD phases towards the achievement of energy efficiency and thermal comfort in tertiary buildings. To do so, the EEPSA (Energy Efficiency Prediction Semantic Assistant) is proposed, which aids data analysts discovering the most relevant variables for the matter at hand, and informing them about relationships among relevant data.
EEPSA leverages Semantic Technologies such as ontologies, ontology-driven rules and ontology-driven data access. More specifically, the EEPSA ontology is the cornerstone of the assistant. This ontology is developed on top of three ODPs (Ontology Design Patterns), which address weaknesses of existing proposals to represent: features of interest and their respective qualities; observations and actuations; the sensors and actuators that generate them; and the procedures used. The ontology is designed so that its customization to address similar problems in different types of buildings can be approached methodically.