Applied Bayesian Networks Rely on Expert Knowledge and Scarce Data Sharing

In this descriptive scoping review, we assessed how Bayesian network structures are built and learned in applied work by screening 5993 recent papers (2020–2025) whose abstracts mention “Bayesian (belief) network” and deeming 3661 relevant. Among these relevant papers, expert knowledge was used in 2059 papers (56.2%): 1785 (48.8%) relied on expert knowledge alone, whereas 274 (7.5%) combined expert input with algorithmic structure learning. Automatic structure learning thus remains underused. Data sharing was scarce: only 129 studies (3.5%) provided functional dataset links, which we curated into an open benchmark index. Among 1106 papers (30.2%) using algorithms without expert knowledge, score-based methods were most common (797, 72.1%; mainly Hill climbing, K2, and tabu), followed by constraint-based methods (194, 17.5%; mainly PC and Grow–Shrink), fixed- or restricted-topology BN classifiers (143, 12.9%; mainly TAN and naive Bayes), and hybrid methods (131, 11.8%; mainly MMHC); bootstrapping appeared in 223 papers (6.1%). Reported practice thus remains concentrated around familiar algorithms.