Table 1.

Descriptions of the key terms in Machine Learning and Big Data AI research approaches [41,42,43,44,45].

Key Term	Description
Knowledge Discovery (KD)	A multiple-step process in data analysis, often managed using CRISP-DM methodology including steps: (1) business understanding; (2) data understanding; (3) data preparation; (4) modelling–decision models generation, patterns extraction; (5) evaluation and (6) deployment-the new knowledge implementation in practice.
Data Mining (DM)	Some experts use it to name the knowledge discovery process. Other experts view data mining as an essential step in the process of knowledge discovery = modelling.
Machine Learning (ML)	The engine within the framework of AI; the collection of techniques allowing computers to undertake complicated tasks by implementation of learning on data (by training and validating the data). The main ML categories are Supervised (SV) Learning, Un-Supervised (USV) Learning and Reinforcement Learning.
The Big Data analytical approach	Enables managing data of the big size and high diversity and complexity; its emergency is due to the rapid advances of high-throughput (-omics) technologies and a wide adoption of eHRs; it is able to challenge the paradigm shift in research on multimorbidity towards the logic of the precision medicine.
Precision medicine	Marked with 4P: Personalized, Predictive, Preventive and Participatory-individualized evaluation and treatments-in contrast to the paradigm “one-size-fits-all”.
The black box concept	Refers to models that use nonlinear transformations to facilitate feature identification; it is used in complex algorithms, such as Artificial Neural Networks (ANN) or a new concept called Deep Learning (DL).