Score-based likelihood ratios for handwriting evidence

Hepler, Amanda B.; Saunders, Christopher P.; Davis, Linda; Buscaglia, JoAnn

doi:10.1016/j.forsciint.2011.12.009

Cited by 93 publications

(109 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Score-based approaches are increasingly popular across multiple branches of forensic science, e.g., [17][18][19][20][21][22][23][24][25]. Quantitative measurements made on objects of interest such as voice recordings, face images, and glass fragments usually result in multivariate data with complex distributions.…”

Section: Score-based Approaches For the Calculation Of Likelihood Ratmentioning

confidence: 99%

“…The second system was a multi-layer perceptron (MLP), three layers with 32 nodes in the hidden (middle) layer. 20 The systems are described in [57]. The protocol for model training and score calculation is described in [56].…”

Section: Forensic Comparison Of Face Imagesmentioning

confidence: 99%

“…15 (MLP scores), and as C llr values in Table 3. Results 20 The MLP system was a closed box with score output scaled to the range −1 to +1.…”

Section: Forensic Comparison Of Face Imagesmentioning

confidence: 99%

See 2 more Smart Citations

Avoiding overstating the strength of forensic evidence: Shrunk likelihood ratios/Bayes factors

Morrison

Poh

2018

Science & Justice

View full text Add to dashboard Cite

A B S T R A C TWhen strength of forensic evidence is quantified using sample data and statistical models, a concern may be raised as to whether the output of a model overestimates the strength of evidence. This is particularly the case when the amount of sample data is small, and hence sampling variability is high. This concern is related to concern about precision. This paper describes, explores, and tests three procedures which shrink the value of the likelihood ratio or Bayes factor toward the neutral value of one. The procedures are: (1) a Bayesian procedure with uninformative priors, (2) use of empirical lower and upper bounds (ELUB), and (3) a novel form of regularized logistic regression. As a benchmark, they are compared with linear discriminant analysis, and in some instances with non-regularized logistic regression. The behaviours of the procedures are explored using Monte Carlo simulated data, and tested on real data from comparisons of voice recordings, face images, and glass fragments.

show abstract

Section: Score-based Approaches For the Calculation Of Likelihood Ratmentioning

confidence: 99%

Section: Forensic Comparison Of Face Imagesmentioning

confidence: 99%

See 1 more Smart Citation

Avoiding overstating the strength of forensic evidence: Shrunk likelihood ratios/Bayes factors

Morrison

Poh

2018

Science & Justice

View full text Add to dashboard Cite

show abstract

“…An increasingly adopted approach [1] uses a logical framework based on Bayes' Theorem to report forensic evidence in terms of likelihood ratios [1,2]. Computer-assisted LR methods (also referred to simply as LR methods), have been developed to assist the forensic practitioner in his role of forensic evaluator [3,4,5,6,7,8,9]. In these methods pattern recognition algorithms are often used for the feature extraction (analysis), the feature comparison, and statistical models are used for the evaluation of the forensic findings.…”

Section: Introductionmentioning

confidence: 99%

Measuring coherence of computer-assisted likelihood ratio methods

Haraksim

Ramos

Meuwly

et al. 2015

Forensic Science International

View full text Add to dashboard Cite

Measuring the performance of forensic evaluation methods that compute likelihood ratios (LRs) is relevant for both the development and the validation of such methods. A framework of performance characteristics categorized as primary and secondary is introduced in this study to help achieve such development and validation. Groundtruth labelled fingerprint data is used to assess the performance of an example likelihood ratio method in terms of those performance characteristics. Discrimination, calibration, and especially the coherence of this LR method are assessed as a function of the quantity and quality of the trace fingerprint data. Assessment of the coherence revealed a weakness of the comparison algorithm in the computer-assisted likelihood ratio method used. ABSTRACTMeasuring the performance of forensic evaluation methods that compute likelihood ratios (LRs) is relevant for both the development and the validation of such methods. A framework of performance characteristics categorized as primary and secondary is introduced in this study to help achieve such development and validation. Ground-truth labelled fingerprint data is used to assess the performance of an example likelihood ratio method in terms of those performance characteristics. Discrimination, calibration, and especially the coherence of this LR method are assessed as a function of the quantity and quality of the trace fingerprint data.Assessment of the coherence revealed a weakness of the comparison algorithm in the computer-assisted likelihood ratio method used.

show abstract

“…It is worth noting that for the models in 65 which features are replaced by e.g. distances between samples (so called scorebased approach [5,6,7]) the rarity refers rather to the frequency of observing a particular distance than a feature.…”

mentioning

confidence: 99%

Hybrid approach combining chemometrics and likelihood ratio framework for reporting the evidential value of spectra

Martyna

Zadora

Neocleous

et al. 2016

Analytica Chimica Acta

View full text Add to dashboard Cite

Hybrid approach combining chemometrics and likelihood ratio framework for reporting the evidential value of spectra. Analytica Chimica Acta, 931, pp. 34-46. (doi:10.1016Acta, 931, pp. 34-46. (doi:10. /j.aca.2016 This is the author's final accepted version.There may be differences between this version and the published version. You are advised to consult the publisher's version if you wish to cite from it.http://eprints.gla.ac.uk/119762/ Chemistry, 3 Ingardena, Poland, 9 Westerplatte, Poland, University of Silesia in Katowice, Institute of Chemistry, Chemometric Research Group, 9 Szkolna, Poland, University of Glasgow, School of Mathematics and Statistics, 15 University Gardens, Glasgow G12 8QW, United Kingdom, Abstract Many chemometric tools are invaluable and have proven effective in data mining and substantial dimensionality reduction of highly multivariate data. This becomes vital for interpreting various physicochemical data due to rapid development of advanced analytical techniques, delivering much information in a single measurement run. This concerns especially spectra, which are frequently used as the subject of comparative analysis in e.g. forensic sciences. In the presented study the microtraces collected from the scenarios of hit-and-run accidents were analysed. Plastic containers and automotive plastics (e.g. bumpers, headlamp lenses) were subjected to Fourier transform infrared spectrometry and car paints were analysed using Raman spectroscopy. In the forensic context analytical results must be interpreted and reported according to the standards of the interpretation schemes acknowledged in forensic sciences using the likelihood ratio approach. However, for proper construction of LR models for highly multivari- * Corresponding authorEmail addresses: rzepecka@chemia.uj.edu.pl (Agnieszka Martyna), gzadora@ies.krakow.pl (Grzegorz Zadora), tereza.neocleous@glasgow.ac.uk (Tereza Neocleous), amichalska@ies.krakow.pl (Aleksandra Michalska), nema.dean@glasgow.ac.uk (Nema Dean) Preprint submitted to Analytica Chimica ActaMarch 25, 2016 ate data, such as spectra, chemometric tools must be employed for substantial data compression. Conversion from classical feature representation to distance representation was proposed for revealing hidden data peculiarities and linear discriminant analysis was further applied for minimising the within-samples variability while maximising the between-samples variability. Both techniques enabled substantial reduction of data dimensionality. Univariate and multivariate likelihood ratio models were proposed for such data. It was shown that the combination of chemometric tools and the likelihood ratio approach is capable of solving the comparison problem of highly multivariate and correlated data after proper extraction of the most relevant features and variance information hidden in the data structure.

show abstract

Score-based likelihood ratios for handwriting evidence

Cited by 93 publications

References 22 publications

Avoiding overstating the strength of forensic evidence: Shrunk likelihood ratios/Bayes factors

Avoiding overstating the strength of forensic evidence: Shrunk likelihood ratios/Bayes factors

Measuring coherence of computer-assisted likelihood ratio methods

Hybrid approach combining chemometrics and likelihood ratio framework for reporting the evidential value of spectra

Contact Info

Product

Resources

About