Fig. 3.

Different success criteria for big versus small data. While multiple methods can be used in each quadrant, to help illustrate, there is a rough mapping to different methods as used in different disciplines. Quadrant A includes techniques such as supervised and unsupervised machine learning, deep learning, reinforcement learning, and recommender systems, commonly used in computer science and the technology industry. Quadrant B includes techniques such as single case experimental designs, N-of-1 cross over designs, and system identification as respectively used in the social and behavioral sciences, medicine, and control systems engineering. Quadrant C includes techniques such as supervised and unsupervised machine learning and deep learning, commonly used in computer science, the technology industry, and various ‘-omics’ efforts. Quadrant D includes techniques articulated as part of the evidence-based pyramid and inferential statistics, commonly used in fields like medicine, epidemiology, public health, and psychology