Abstract:Bayesian networks (BNs) are a kind of graphical model that formally combines elements of graph and probability theory. BNs are a mathematically and statistically rigorous technique allowing their user to define a pictorial representation of assumed dependencies and influences among a set of variables deemed to be relevant for a particular inferential problem. The formalism allows one to process newly acquired evidence according to the rules of probability calculus. Applications of BNs have been reported in various forensic disciplines. However, there seems to be some reluctance to consider BNs as a more general framework for representing and evaluating sources of uncertainties associated with scientific evidence. Notably, BNs are widely thought of as an essentially numerical method, requiring "exact" numbers with a high "accuracy". The present paper aims to draw the reader's attention to the point that the availability of hard numerical data is not a necessary requirement for using BNs in forensic science. An abstraction of quantitative BNs, known as qualitative probabilistic networks (QPNs), and sensitivity analyses are presented and their potential applications discussed. As a main difference to their quantitative counterpart, QPNs contain qualitative probabilistic relationships instead of numerical relations. Sensitivity analyses consist of varying the probabilities assigned to one or more variables and evaluating the effect on one or more other variables of interest. Both QPNs and sensitivity analyses appear to be useful concepts that permit one to work in contexts with acute lack of numerical data and where reasoning consistent with the laws of probability should nevertheless be performed.

Quantitative Combination of Different Bayesian Networks

An automatic adaptive method to combine summary statistics in approximate Bayesian computation

Quantized Distributed Estimation with Adaptive Combiner

Fusing Uncertain Knowledge In Time-Series Data

Extraction of Belief Knowledge from a Relational Database for Quantitative Bayesian Network Inference

The Quantified Boolean Bayesian Network: Theory and Experiments with a Logical Graphical Model

Class dependent feature scaling method via restrictive Bayesian network classifier combination

Decomposition and Approximation of Loopy Bayesian Networks.

Learning the Structure of Bayesian Networks: A Quantitative Assessment of the Effect of Different Algorithmic Schemes

Research on the Combihation of Bayesian Learning and Reinforcement Learning

Extracting Bayesian networks from multiple copies of a quantum system

Combination of Upper and Lower Probabilities

Multiple Independent Quantum States Sharing under Collaboration of Agents in Quantum Networks

Continuity approximation in hybrid Bayesian networks structure learning

A New Hybrid Method for Learning Bayesian Networks: Separation and Reunion.

Reductions of discrete Bayesian networks via lumping

Probabilistic graphs using coupled random variables

Bayesian networks and probabilistic reasoning about scientific evidence when there is a lack of data

Bayesian network with quantitative input for maritime risk analysis

Quantifying conditional probability tables in Bayesian networks: Bayesian regression for scenario-based encoding of elicited expert assessments on feral pig habitat

Quotients of Probabilistic Boolean Networks