*2.2. Big Data*

Traditionally big data refers to the concept of an amount of data that exceeds the capacity of conventional software to be captured, managed, and processed in a reasonable amount of time. Nowadays, the concept extends to analyzing user behavior, extracting value from stored data, and formulating predictions through the patterns observed. A first approximation to this definition was given in 2012 by [37].

Big data uses the following characteristics described by the three V's: volume, variety, and velocity, and several other characteristics including veracity, value, and the identification of nonlinear systems (from large data sets) to reveal relationships or to make predictions of outcomes and behaviors [38–40].

From the large amount of data generated in a smart home, whether internal (through the IoT network integrated into the home) or external (such as weather or electricity prices), it would be interesting to improve energy efficiency, to analyze these data and extract all the relevant information they can provide to the system, so the use of big data can be an essential tool, providing grea<sup>t</sup> value for the optimization of home resources and user comfort. However, the storage, processing, and analysis of this large volume of continuously generated data, while maintaining their security and privacy, is a significant challenge for HEMS.

To this end, HERMES uses various strategies to process volumes of data, store the information periodically and in real-time, and process it to obtain an analysis and projection of the data to trigger specific automated actions without user intervention. It also offers information that is provided to the user through the Expert Assistant to guide decision making or as information on predictions or patterns detected through machine learning (ML).
