The likelihood ratios calculated using the novel PCA approach were compared with previously reported LRs calculated using a more computationally intensive MultivariateKernel (MVK) model, followed by a calibration step using a Pool Adjacent Violators (PAV) algorithm.
Dimensionality reduction of multivariate elemental concentrations of glass is reported for computing likelihood ratios (LRs). The LRs calculated using principal component analysis (PCA) and a post hoc calibration steps result in low (<1 percent) false inclusions when comparing glass samples known to originate from different sources and low (<1 percent) false exclusions when comparing glass samples known to originate from the same source. The current study found that in both cases, the calibrated LRs limited the magnitude of the misleading evidence, providing only weak to moderate support for the incorrect hypotheses. Most of the different pairs that were found to be falsely included were explained by chemical relatedness (same manufacturer of the glass sources in very close time interval between manufacture). The computation of LRs using dimensionality reduction of elemental concentrations using PCA may transfer to other multivariate data-generating evidence types. (publisher abstract modified)
Downloads
Similar Publications
- A Multi-Stream Fusion Approach with One-Class Learning for Audio-Visual Deepfake Detection
- National Institute of Justice Fiscal Year 2023 Annual Report
- Enhancing Fault Ride-Through Capacity of DFIG-Based WPs by Adaptive Backstepping Command Using Parametric Estimation in Non-Linear Forward Power Controller Design