Evaluating evidence in linked crimes with multiple offenders

A collection of idioms for modeling activity level evaluations in forensic science

“…One may also use separate BNs for each hypothesis. Cause-consequence idioms are frequently presented in forensic and legal research, whether in template models [ 3 , 9 , 13 , 26 , 27 , [37] , [38] , [39] , [40] , [41] , [42] , [43] , [44] , [45] , [46] , [47] , [48] , [49] , [50] , [51] , [52] , [53] , [54] , [55] , [56] , [57] , [58] , [59] , [60] , [61] , [62] , [63] ] or case models [ [16] , [17] , [18] , [19] , [64] , [65] , [66] , [67] , [68] , [69] , [70] , [71] , [72] , [73] , [74] , [75] , [76] ]. …”

Evaluation of forensic genetics findings given activity level propositions: A review

“…Features or factors to consider Examples of evaluations at this level in the hierarchy [11,[19][20][21] [ 6,8,19,[22][23][24][25][26][27][28][29] [ [30][31][32] [4, [33][34][35] [18]…”

“…There have been numerous applications of BN within forensic science ranging from quality control monitoring [53], preparation for legal challenges [54], complex pedigree evaluation [55][56][57], to DNA profile mixture evaluation [58], helping to address activity level propositions [6] (see, e.g., [48] for a review) and for crime level investigation [21,59]. We direct the reader to [60] for explanations of the structure and terminology of BNs.…”

An interactive human centered data science approach towards crime pattern analysis

“…[32] utilized Bayesian networks to link evidence in crimes. In another research [33] has utilized Bayesian networks for modeling multiple offenders for two separate offenses. Another example is from Reference [34] who employed crime date, location, and the criminal name and criminals acquaintances as clues to predict the posterior probability of a criminal to be associated towards an unsolved crime.…”