似然比方法体系在法庭科学中的研究进展
|
摘要
似然比自提出以来,经过法庭科学家和统计学家的不断完善与创新,已经形成了大量理论成果。在将似然比的理论成果进行应用的过程中,国外法庭科学家又不断改进模型从而使得似然比植根于法庭科学的各个分支。似然比的引入使得法庭科学学科根基更加牢固,内涵更加丰富,也为法庭证据证明力的量化评估提供了途径。本文在综合国外大量关于似然比研究文献的过程中,着重分析了似然比在法化学、指纹、枪弹痕迹鉴定中的应用与研究,介绍了两种似然比的计算方法(特征变量型似然比、评分法似然比)以及似然比模型的几种评价方式(错误率、Tippett图等)。以期推动国内关于似然比的理论与实证研究。
Since the likelihood ratio method was proposed, many theoretical achievements have been made through continuous efforts and innovations from collaborations between forensic scientists and statisticians. In application of the likelihood ratio method, foreign forensic scientists continue to improve the model to make it well-adapted to every sub-divisions of forensic science. The introduction of the likelihood ratio strengthened the foundation of forensic science, enriched its content, and provided an approach for testing the quantitative evaluation of the probative force of forensic evidence. This paper introduces two kinds of likelihood ratio calculation methods(the feature-based likelihood ratio method and score-based likelihood ratio) and several different ways to evaluate the performance of likelihood ratio models(such as error rate, Tippett plot, etc.), with special emphasis on the analysis of the application of likelihood ratio method in forensic chemistry, fi ngerprint identification and fi rearm identification on the basis of substantial related studies in other countries, aiming to promote China's theoretical and empirical studies on the likelihood ratio method.
Since the likelihood ratio method was proposed, many theoretical achievements have been made through continuous efforts and innovations from collaborations between forensic scientists and statisticians. In application of the likelihood ratio method, foreign forensic scientists continue to improve the model to make it well-adapted to every sub-divisions of forensic science. The introduction of the likelihood ratio strengthened the foundation of forensic science, enriched its content, and provided an approach for testing the quantitative evaluation of the probative force of forensic evidence. This paper introduces two kinds of likelihood ratio calculation methods(the feature-based likelihood ratio method and score-based likelihood ratio) and several different ways to evaluate the performance of likelihood ratio models(such as error rate, Tippett plot, etc.), with special emphasis on the analysis of the application of likelihood ratio method in forensic chemistry, fi ngerprint identification and fi rearm identification on the basis of substantial related studies in other countries, aiming to promote China's theoretical and empirical studies on the likelihood ratio method.
引文
[1]Ali T,Spreeuwers L,Veldhuis R.Computation of likelihood ratio from small sample set of within-source variability[J].European Academy of Forensic Science,2012.
[2]Ishihara,Shunichi.Strength of linguistic text evidence:A fused forensic text comparison system[J].Forensic Science International,2017:S0379073817302475.
[3]Hepler A B,Saunders C P,Davis L J,et al.Score-based likelihood ratios for handwriting evidence[J].Forensic Science International,2012,219(1-3):129-140.
[4]Chen X H,Champod C,Yang X,et al.Assessment of signature handwriting evidence via score-based likelihood ratio based on comparative measurement of relevant dynamic features.[J].Forensic Science International,2018,282:101.
[5]Benjamin B P D,Anurag J M S,Sung J M S,et al.A Statistical Validation of the Individuality and Repeatability of Striated Tool Marks:Screwdrivers and Tongue and Groove Pliers[J].Journal of Forensic Sciences,2010,55(2):348-57.
[6]张翠玲,谭铁君.基于贝叶斯统计推理的法庭证据评价[J].刑事技术,2018,43(04):265-271.
[7]Evett I W.A Bayesian Approach to the Problem of Interpreting Glass Evidence in Forensic Science Casework[J].Journal of the Forensic Science Society,1986,26(1):3-18.
[8]Evett I W,Cage P E,Aitken C G G.Evaluation of the Likelihood Ratio for Fibre Transfer Evidence in Criminal Cases[J].Journal of the Royal Statistical Society,1987,36(2):174-180.
[9]C.G.G.Aitken,D.Lucy.Evaluation of trace evidence in the form of multivariate data[J].Journal of the Royal Statistical Society:Series C(Applied Statistics),2004,53(1).
[10]Aitken C,Gold E.Evidence evaluation for discrete data-Forensic Science International[J].Forensic Science International,2013,230(1-3):147-155.
[11]Zadora G,Neocleous T.Likelihood ratio model for classifi cation of forensic evidence[J].Analytica Chimica Acta,2009,642(1-2):266-278.
[12]Zadora G.Glass analysis for forensic purposes-a comparison of classifi cation methods[J].Journal of Chemometrics,2007,21(5‐6):174-186.
[13]Zadora G,Martyna A,Ramos D,et al.Statistical Analysis in Forensic Science:Evidential Value of Multivariate Physicochemical Data[M]//Encyclopedia of Analytical Science.2005:373-378.
[14]Taylor D,Hicks T,Champod C.Using sensitivity analyses in Bayesian Networks to highlight the impact of data paucity and direct future analyses:a contribution to the debate on measuring and reporting the precision of likelihood ratios[J].Science&Justice,2016,56(5):402-410.
[15]Champod C,Evett I W,Jackson G.Establishing the most appropriate databases for addressing source level propositions[J].Science&Justice Journal of the Forensic Science Society,2004,44(3):153-164.
[16]Ali T,Spreeuwers L,Veldhuis R,et al.Sampling variability in forensic likelihood-ratio computation:A simulation study[J].Science&Justice,2015,55(6):499-508.
[17]Taylor D,Bright J A,Buckleton J,et al.An illustration of the effect of various sources of uncertainty on DNA likelihood ratio calculations[J].Forensic Science International Genetics,2014,11(1):56.
[18]Langenberg G.A Performance study of the ACE-V process:A pilot study to measure the accuracy,precision,reproducibility,repeatability,and biasability of conclusions resulting from the ACE-V process[J].Journal of Forensic Identifi cation,2009,59(2):219-257.
[19]Saks MJ.Forensic identifi cation:From a faith-based"Science"to a scientifi c science[J].Forensic Science International,2010,201(1-3):14.
[20]Mnookin J L.The Courts,the NAS,and the Future of Forensic Science[J].Brooklyn Law Review,2010.
[21]Ulery B T,Hicklin R A,Buscaglia J,et al.Accuracy and reliability of forensic latent fi ngerprint decisions[J].Proceedings of the National Academy of Sciences of the United States of America,2011,108(19):7733.
[22]Champod C.Friction Ridge Examination(Fingerprints):Interpretation of[M]//Wiley Encyclopedia of Forensic Science.John Wiley&Sons,Ltd,2009.
[23]Evett I W,Jackson G,Lambert J A,et al.The impact of the principles of evidence interpretation on the structure and content of statements[J].Science&Justice Journal of the Forensic Science Society,2000,40(4):233.
[24]Al-Haddad L M,Neumann C.Benefi ts and challenges of the use of fi ngerprint statistical models in casework[J].Science&Justice,2010,50(1):32-33.
[25]Egli Anthonioz N M,Champod C.Evidence evaluation in fingerprint comparison and automated fingerprint identifi cation systems--Modeling between fi nger variability.[J].Forensic Science International,2014,167(2):189-195.
[26]International Association for Identifi cation.IAI Resolution 2010-18,2010;http://www.swgfast.org/Resources/100716_IAI_Resolution_2010-18.pdf(accessed August 1,2016).
[27]Neumann C,Champod C,Puchsolis R,et al.Computation of likelihood ratios in fingerprint identification for confi gurations of three minutiae.[J].Journal of Forensic Sciences,2007,52(1):54-64.
[28]Neumann C,Evett I W,Skerrett J.Quantifying the weight of evidence from a forensic fi ngerprint comparison:a new paradigm[J].Journal of the Royal Statistical Society,2012,175(2):371-415.
[29]Neumann C,Evett I W,Skerrett J E,et al.Quantitative assessment of evidential weight for a fi ngerprint comparison I.Generalisation to the comparison of a mark with set of ten prints from a suspect.[J].Forensic Science International,2011,207(1-3):101-105.
[30]Leegwater A J,Meuwly D,Sjerps M,et al.Performance Study of a Score-based Likelihood Ratio System for Forensic Fingermark Comparison.[J].Journal of Forensic Sciences,2017,62(3):626.
[31]Evett I W,Jackson G,Lambert J A,et al.The impact of the principles of evidence interpretation on the structure and content of statements[J].Science&Justice,2000,40(4):233.
[32]Bolck A,Dstoel R,Amp I A,et al.LR models for evidence evaluation[J].Chinese Journal of Forensic Sciences,2012,4:28-42.
[33]Broeders A P A.Some observations on the use of probability scales in forensic identifi cation[J].International Journal of Speech Language and the Law,2007,6(2):228-241.
[34]Meuwly D,Veldhuis R.Forensic biometrics:From two communities to one discipline[C]//Biometrics Special Interest Group.IEEE,2012:1-12.
[35]Kirk J N,Law E F,Morris K B.Estimation of changes in breech face and fi ring pin marks over consecutive discharges and its impact on an IBIS?,Heritage?System[J].Forensic Science International,2017,278:47-51.
[36]Ogunc GI,Oralli A,Kara H,Sayar A,Sakarya U,&Arican YE.The performance of BALISTIKA 2010 system for 9 mm×19 mm and 7.65 mm×17 mm cartridge case correlation.[J].Forensic Science International,2013,232(1-3):104-110.
[37]Rahm J.Evaluation of an electronic comparison system and implementation of a quantitative effectiveness criterion[J].Forensic Science International,2012,214(1-3):173.
[38]FTI,Inc,Integrated Ballistics Identifi cation System Training Guide,(2002).
[39]Kinder J D.Ballistic fi ngerprinting databases[J].Science&Justice,2002,42(4):197-203.
[40]De Ceuster J,Dujardin S.The reference ballistic imaging database revisited[J].Forensic Science International,2015,248:82-87.
[41]Kinder J D,Tulleners F,Thiebaut H.Reference ballistic imaging database performance[J].Forensic Science International,2004,140(2-3):207-215.
[42]Song J,Vorburger T V,Chu W,et al.Estimating Error Rates for Firearm Evidence Identifi cations in Forensic Science[J].Forensic Science International,2017,284.Fabiano R P D,Sc R H B,Sc E M M,et al.Objective Evaluation of Subclass Characteristics on Breech Face Marks[J].Journal of Forensic Sciences,2017,62(2):417.
[43]Fabiano R P D,Sc R H B,Sc E M M,et al.Objective Evaluation of Subclass Characteristics on Breech Face Marks[J].Journal of Forensic Sciences,2017,62(2):417.
[44]Riva F,Champod C.Automatic comparison and evaluation of impressions left by a fi rearm on fi red cartridge cases[J].Journal of Forensic Sciences,2014,59(3):637-647.
[45]Nichols R G,B.S.Firearm and toolmark identification criteria:A review of the literature[J].Journal of Forensic Sciences,1997,48(2):318-27.
[46]Nichols R G.Firearm and toolmark identification criteria:A review of the literature,part II[J].Journal of Forensic Sciences,2003,48(2):318-27.
[47]https://afte.org/about-us/what-is-afte/afte-theory-of-identifi cation
[48]L?vblad K O,Baird A E.A probabilistic approach to the joint evaluation of fi rearm evidence and gunshot residues[J].Forensic Science International,2005,163(1):18-33.
[49]Kerkhoff W,Stoel R D,Mattijssen E,et al.The likelihood ratio approach in cartridge case and bullet comparison[J].AFTE J,2013,45(3):284-289.
[50]Bunch S,Wevers G.Application of likelihood ratios for fi rearm and toolmark analysis.[J].Science&Justice Journal of the Forensic Science Society,2013,53(2):223-229.
[51]Ramos D,Gonzalezrodriguez J.Reliable support:Measuring calibration of likelihood ratios[J].Forensic Science International,2013,230(1-3):156-169.
[52]Morrison,Stewart G.Special issue on measuring and reporting the precision of forensic likelihood ratios:Introduction to the debate[J].Science&Justice,2016:S1355030616300260.
[53]Bunch S G,Ph.D.Consecutive matching striation criteria:a general critique[J].Journal of Forensic Sciences,2000,45(5):955.
[2]Ishihara,Shunichi.Strength of linguistic text evidence:A fused forensic text comparison system[J].Forensic Science International,2017:S0379073817302475.
[3]Hepler A B,Saunders C P,Davis L J,et al.Score-based likelihood ratios for handwriting evidence[J].Forensic Science International,2012,219(1-3):129-140.
[4]Chen X H,Champod C,Yang X,et al.Assessment of signature handwriting evidence via score-based likelihood ratio based on comparative measurement of relevant dynamic features.[J].Forensic Science International,2018,282:101.
[5]Benjamin B P D,Anurag J M S,Sung J M S,et al.A Statistical Validation of the Individuality and Repeatability of Striated Tool Marks:Screwdrivers and Tongue and Groove Pliers[J].Journal of Forensic Sciences,2010,55(2):348-57.
[6]张翠玲,谭铁君.基于贝叶斯统计推理的法庭证据评价[J].刑事技术,2018,43(04):265-271.
[7]Evett I W.A Bayesian Approach to the Problem of Interpreting Glass Evidence in Forensic Science Casework[J].Journal of the Forensic Science Society,1986,26(1):3-18.
[8]Evett I W,Cage P E,Aitken C G G.Evaluation of the Likelihood Ratio for Fibre Transfer Evidence in Criminal Cases[J].Journal of the Royal Statistical Society,1987,36(2):174-180.
[9]C.G.G.Aitken,D.Lucy.Evaluation of trace evidence in the form of multivariate data[J].Journal of the Royal Statistical Society:Series C(Applied Statistics),2004,53(1).
[10]Aitken C,Gold E.Evidence evaluation for discrete data-Forensic Science International[J].Forensic Science International,2013,230(1-3):147-155.
[11]Zadora G,Neocleous T.Likelihood ratio model for classifi cation of forensic evidence[J].Analytica Chimica Acta,2009,642(1-2):266-278.
[12]Zadora G.Glass analysis for forensic purposes-a comparison of classifi cation methods[J].Journal of Chemometrics,2007,21(5‐6):174-186.
[13]Zadora G,Martyna A,Ramos D,et al.Statistical Analysis in Forensic Science:Evidential Value of Multivariate Physicochemical Data[M]//Encyclopedia of Analytical Science.2005:373-378.
[14]Taylor D,Hicks T,Champod C.Using sensitivity analyses in Bayesian Networks to highlight the impact of data paucity and direct future analyses:a contribution to the debate on measuring and reporting the precision of likelihood ratios[J].Science&Justice,2016,56(5):402-410.
[15]Champod C,Evett I W,Jackson G.Establishing the most appropriate databases for addressing source level propositions[J].Science&Justice Journal of the Forensic Science Society,2004,44(3):153-164.
[16]Ali T,Spreeuwers L,Veldhuis R,et al.Sampling variability in forensic likelihood-ratio computation:A simulation study[J].Science&Justice,2015,55(6):499-508.
[17]Taylor D,Bright J A,Buckleton J,et al.An illustration of the effect of various sources of uncertainty on DNA likelihood ratio calculations[J].Forensic Science International Genetics,2014,11(1):56.
[18]Langenberg G.A Performance study of the ACE-V process:A pilot study to measure the accuracy,precision,reproducibility,repeatability,and biasability of conclusions resulting from the ACE-V process[J].Journal of Forensic Identifi cation,2009,59(2):219-257.
[19]Saks MJ.Forensic identifi cation:From a faith-based"Science"to a scientifi c science[J].Forensic Science International,2010,201(1-3):14.
[20]Mnookin J L.The Courts,the NAS,and the Future of Forensic Science[J].Brooklyn Law Review,2010.
[21]Ulery B T,Hicklin R A,Buscaglia J,et al.Accuracy and reliability of forensic latent fi ngerprint decisions[J].Proceedings of the National Academy of Sciences of the United States of America,2011,108(19):7733.
[22]Champod C.Friction Ridge Examination(Fingerprints):Interpretation of[M]//Wiley Encyclopedia of Forensic Science.John Wiley&Sons,Ltd,2009.
[23]Evett I W,Jackson G,Lambert J A,et al.The impact of the principles of evidence interpretation on the structure and content of statements[J].Science&Justice Journal of the Forensic Science Society,2000,40(4):233.
[24]Al-Haddad L M,Neumann C.Benefi ts and challenges of the use of fi ngerprint statistical models in casework[J].Science&Justice,2010,50(1):32-33.
[25]Egli Anthonioz N M,Champod C.Evidence evaluation in fingerprint comparison and automated fingerprint identifi cation systems--Modeling between fi nger variability.[J].Forensic Science International,2014,167(2):189-195.
[26]International Association for Identifi cation.IAI Resolution 2010-18,2010;http://www.swgfast.org/Resources/100716_IAI_Resolution_2010-18.pdf(accessed August 1,2016).
[27]Neumann C,Champod C,Puchsolis R,et al.Computation of likelihood ratios in fingerprint identification for confi gurations of three minutiae.[J].Journal of Forensic Sciences,2007,52(1):54-64.
[28]Neumann C,Evett I W,Skerrett J.Quantifying the weight of evidence from a forensic fi ngerprint comparison:a new paradigm[J].Journal of the Royal Statistical Society,2012,175(2):371-415.
[29]Neumann C,Evett I W,Skerrett J E,et al.Quantitative assessment of evidential weight for a fi ngerprint comparison I.Generalisation to the comparison of a mark with set of ten prints from a suspect.[J].Forensic Science International,2011,207(1-3):101-105.
[30]Leegwater A J,Meuwly D,Sjerps M,et al.Performance Study of a Score-based Likelihood Ratio System for Forensic Fingermark Comparison.[J].Journal of Forensic Sciences,2017,62(3):626.
[31]Evett I W,Jackson G,Lambert J A,et al.The impact of the principles of evidence interpretation on the structure and content of statements[J].Science&Justice,2000,40(4):233.
[32]Bolck A,Dstoel R,Amp I A,et al.LR models for evidence evaluation[J].Chinese Journal of Forensic Sciences,2012,4:28-42.
[33]Broeders A P A.Some observations on the use of probability scales in forensic identifi cation[J].International Journal of Speech Language and the Law,2007,6(2):228-241.
[34]Meuwly D,Veldhuis R.Forensic biometrics:From two communities to one discipline[C]//Biometrics Special Interest Group.IEEE,2012:1-12.
[35]Kirk J N,Law E F,Morris K B.Estimation of changes in breech face and fi ring pin marks over consecutive discharges and its impact on an IBIS?,Heritage?System[J].Forensic Science International,2017,278:47-51.
[36]Ogunc GI,Oralli A,Kara H,Sayar A,Sakarya U,&Arican YE.The performance of BALISTIKA 2010 system for 9 mm×19 mm and 7.65 mm×17 mm cartridge case correlation.[J].Forensic Science International,2013,232(1-3):104-110.
[37]Rahm J.Evaluation of an electronic comparison system and implementation of a quantitative effectiveness criterion[J].Forensic Science International,2012,214(1-3):173.
[38]FTI,Inc,Integrated Ballistics Identifi cation System Training Guide,(2002).
[39]Kinder J D.Ballistic fi ngerprinting databases[J].Science&Justice,2002,42(4):197-203.
[40]De Ceuster J,Dujardin S.The reference ballistic imaging database revisited[J].Forensic Science International,2015,248:82-87.
[41]Kinder J D,Tulleners F,Thiebaut H.Reference ballistic imaging database performance[J].Forensic Science International,2004,140(2-3):207-215.
[42]Song J,Vorburger T V,Chu W,et al.Estimating Error Rates for Firearm Evidence Identifi cations in Forensic Science[J].Forensic Science International,2017,284.Fabiano R P D,Sc R H B,Sc E M M,et al.Objective Evaluation of Subclass Characteristics on Breech Face Marks[J].Journal of Forensic Sciences,2017,62(2):417.
[43]Fabiano R P D,Sc R H B,Sc E M M,et al.Objective Evaluation of Subclass Characteristics on Breech Face Marks[J].Journal of Forensic Sciences,2017,62(2):417.
[44]Riva F,Champod C.Automatic comparison and evaluation of impressions left by a fi rearm on fi red cartridge cases[J].Journal of Forensic Sciences,2014,59(3):637-647.
[45]Nichols R G,B.S.Firearm and toolmark identification criteria:A review of the literature[J].Journal of Forensic Sciences,1997,48(2):318-27.
[46]Nichols R G.Firearm and toolmark identification criteria:A review of the literature,part II[J].Journal of Forensic Sciences,2003,48(2):318-27.
[47]https://afte.org/about-us/what-is-afte/afte-theory-of-identifi cation
[48]L?vblad K O,Baird A E.A probabilistic approach to the joint evaluation of fi rearm evidence and gunshot residues[J].Forensic Science International,2005,163(1):18-33.
[49]Kerkhoff W,Stoel R D,Mattijssen E,et al.The likelihood ratio approach in cartridge case and bullet comparison[J].AFTE J,2013,45(3):284-289.
[50]Bunch S,Wevers G.Application of likelihood ratios for fi rearm and toolmark analysis.[J].Science&Justice Journal of the Forensic Science Society,2013,53(2):223-229.
[51]Ramos D,Gonzalezrodriguez J.Reliable support:Measuring calibration of likelihood ratios[J].Forensic Science International,2013,230(1-3):156-169.
[52]Morrison,Stewart G.Special issue on measuring and reporting the precision of forensic likelihood ratios:Introduction to the debate[J].Science&Justice,2016:S1355030616300260.
[53]Bunch S G,Ph.D.Consecutive matching striation criteria:a general critique[J].Journal of Forensic Sciences,2000,45(5):955.