Combined interpretation of objective firearm evidence comparison algorithms using Bayesian network.
bulletAn object of class bn.fit. Refer to the documentation of bnlearn for details.
A discrete Bayesian network to leverage the strengths of individual approaches to evaluate the similarity of features on two bullets. The network was available in a repository. The vertices are:
(NotSource, Source);
Cross-correlation function (CCF_0_1, CCF_1_2, CCF_2_3, CCF_3_4, CCF_4_5, CCF_5_6, CCF_6_7, CCF_7_8, CCF_8_9, CCF_9_10);
Congruent matching profile segments (CMPS_0, CMPS_1, CMPS_2, CMPS_3, ... , CMPS_27);
Random forest (RF_0_1, RF_1_2, RF_2_3, RF_3_4, RF_4_5, RF_5_6, RF_6_7, RF_7_8, RF_8_9, RF_9_10);
Consecutively matching striae (CMS_0, CMS_1, .... , CMS_29);
@usage NULL
Spaulding, J. S., & LaCasse, L. S. (2024). Combined interpretation of objective firearm evidence comparison algorithms using Bayesian networks. Journal of Forensic Sciences.