This article studies expected performance and practical feasibility of the most commonly used classes of source-level likelihood-ratio (LR) systems when applied to a trace–reference comparison problem. The article compares performance of these classes of LR systems (used to update prior odds) to each other and to the use of prior odds only, using strictly proper scoring rules as performance measures. It also explores practical feasibility of the classes of LR systems. The present analysis allows for a ranking of these classes of LR systems: from specific-source feature-based to common-source anchored or non-anchored score-based. A trade-off between performance and practical feasibility is observed, meaning that the best performing class of LR systems is the hardest to realize in practice, while the least performing class is the easiest to realize in practice. The other classes of LR systems are in between the two extremes. The one positive exception is a common-source feature-based LR system, with good performance and relatively low experimental demands. This article also argues against the claim that some classes of LR systems should not be used, by showing that all systems have merit (when updating prior odds) over just using the prior odds (i.e. not using the LR system).
{"title":"From specific-source feature-based to common-source score-based likelihood-ratio systems: ranking the stars","authors":"P. Vergeer","doi":"10.1093/lpr/mgad005","DOIUrl":"https://doi.org/10.1093/lpr/mgad005","url":null,"abstract":"\u0000 This article studies expected performance and practical feasibility of the most commonly used classes of source-level likelihood-ratio (LR) systems when applied to a trace–reference comparison problem. The article compares performance of these classes of LR systems (used to update prior odds) to each other and to the use of prior odds only, using strictly proper scoring rules as performance measures. It also explores practical feasibility of the classes of LR systems. The present analysis allows for a ranking of these classes of LR systems: from specific-source feature-based to common-source anchored or non-anchored score-based. A trade-off between performance and practical feasibility is observed, meaning that the best performing class of LR systems is the hardest to realize in practice, while the least performing class is the easiest to realize in practice. The other classes of LR systems are in between the two extremes. The one positive exception is a common-source feature-based LR system, with good performance and relatively low experimental demands. This article also argues against the claim that some classes of LR systems should not be used, by showing that all systems have merit (when updating prior odds) over just using the prior odds (i.e. not using the LR system).","PeriodicalId":48724,"journal":{"name":"Law Probability & Risk","volume":null,"pages":null},"PeriodicalIF":0.7,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46071750","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper we model the criminal standard of proof as a twofold standard requiring sufficient probability of the factum probandum and sufficient informativeness. The focus of the paper is on the latter requirement, and we use decision theory to develop a model for sufficient informativeness. We demonstrate that sufficient informativeness is fundamentally a question of information economics and switch-ability. In our model, sufficient informativeness is a cost-benefit-analysis of further investigations that involves a prediction of the possibility that such investigations will produce evidence that switches the decision from conviction to acquittal. Critics of the Bayesian approach to legal evidence have claimed that ‘weight’ cannot be captured in a Bayesian model. Contrary to this claim, our model shows how sufficient informativeness can be modelled as a second order probability.
{"title":"Information economics in the criminal standard of proof","authors":"Christian Dahlman, A. Nordgaard","doi":"10.1093/lpr/mgad004","DOIUrl":"https://doi.org/10.1093/lpr/mgad004","url":null,"abstract":"\u0000 In this paper we model the criminal standard of proof as a twofold standard requiring sufficient probability of the factum probandum and sufficient informativeness. The focus of the paper is on the latter requirement, and we use decision theory to develop a model for sufficient informativeness. We demonstrate that sufficient informativeness is fundamentally a question of information economics and switch-ability. In our model, sufficient informativeness is a cost-benefit-analysis of further investigations that involves a prediction of the possibility that such investigations will produce evidence that switches the decision from conviction to acquittal. Critics of the Bayesian approach to legal evidence have claimed that ‘weight’ cannot be captured in a Bayesian model. Contrary to this claim, our model shows how sufficient informativeness can be modelled as a second order probability.","PeriodicalId":48724,"journal":{"name":"Law Probability & Risk","volume":null,"pages":null},"PeriodicalIF":0.7,"publicationDate":"2023-03-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48863248","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Odds ratios have several advantages over other methods of measuring the degree of under-representation of cognizable classes of potential jurors. In particular, its advantage over comparative disparity is that it does not measure the extent of under-representation of some groups against an aggregate that includes the very group in question. Odds ratios in jury analysis are directly interpretable as the factor by which one’s probability of being on the jury is advanced or diminished by membership in a specified group.
{"title":"Odds ratios as a measure of disproportionate treatment: application to jury venires","authors":"J. Kadane","doi":"10.1093/lpr/mgad003","DOIUrl":"https://doi.org/10.1093/lpr/mgad003","url":null,"abstract":"\u0000 Odds ratios have several advantages over other methods of measuring the degree of under-representation of cognizable classes of potential jurors. In particular, its advantage over comparative disparity is that it does not measure the extent of under-representation of some groups against an aggregate that includes the very group in question. Odds ratios in jury analysis are directly interpretable as the factor by which one’s probability of being on the jury is advanced or diminished by membership in a specified group.","PeriodicalId":48724,"journal":{"name":"Law Probability & Risk","volume":null,"pages":null},"PeriodicalIF":0.7,"publicationDate":"2023-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48972191","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Inconclusive conclusions in forensic science: rejoinders to Scurich, Morrison, Sinha and Gutierrez","authors":"H. Arkes, J. Koehler","doi":"10.1093/lpr/mgad002","DOIUrl":"https://doi.org/10.1093/lpr/mgad002","url":null,"abstract":"","PeriodicalId":48724,"journal":{"name":"Law Probability & Risk","volume":null,"pages":null},"PeriodicalIF":0.7,"publicationDate":"2023-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48377548","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Signal detection theory fails to account for real-world consequences of inconclusive decisions","authors":"Maneka Sinha, Richard Gutíerrez","doi":"10.1093/lpr/mgad001","DOIUrl":"https://doi.org/10.1093/lpr/mgad001","url":null,"abstract":"","PeriodicalId":48724,"journal":{"name":"Law Probability & Risk","volume":null,"pages":null},"PeriodicalIF":0.7,"publicationDate":"2023-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42227176","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This is Part II of a novel ‘epistemic’ theory of the criminal process. Part I, a formal treatment of how poor measurements degrade control over systems of classification, was published in the last issue. This distinguished four distinct realms of control a policymaker may inhabit and established that the criminal process in the USA is trapped in the third best of these, routinely unable to establish that any policy is better than any other. Here, I first use the new epistemic concepts to mathematize Packer’s influential ‘two models’ theory, both a validation of the formal work and a way of giving it traction in the real world. I then use the new techniques to critique Posner’s law and economics and reject it on methodological grounds. I then argue that the activity of making policy for the criminal process, including any quantativity, is largely a form of ritual. The function of this has been to keep at bay an enduring pressure for rationality, as in Weber’s rationalization thesis, that could not be satisfied. Now that we have the capacity for better measurements, a rational approach is for the first time attainable, and I contemplate the end of the old order.
{"title":"An epistemic theory of the criminal process, Part II: Packer, Posner and epistemic pressure","authors":"William Cullerne Bown","doi":"10.1093/lpr/mgac014","DOIUrl":"https://doi.org/10.1093/lpr/mgac014","url":null,"abstract":"\u0000 This is Part II of a novel ‘epistemic’ theory of the criminal process. Part I, a formal treatment of how poor measurements degrade control over systems of classification, was published in the last issue. This distinguished four distinct realms of control a policymaker may inhabit and established that the criminal process in the USA is trapped in the third best of these, routinely unable to establish that any policy is better than any other. Here, I first use the new epistemic concepts to mathematize Packer’s influential ‘two models’ theory, both a validation of the formal work and a way of giving it traction in the real world. I then use the new techniques to critique Posner’s law and economics and reject it on methodological grounds. I then argue that the activity of making policy for the criminal process, including any quantativity, is largely a form of ritual. The function of this has been to keep at bay an enduring pressure for rationality, as in Weber’s rationalization thesis, that could not be satisfied. Now that we have the capacity for better measurements, a rational approach is for the first time attainable, and I contemplate the end of the old order.","PeriodicalId":48724,"journal":{"name":"Law Probability & Risk","volume":null,"pages":null},"PeriodicalIF":0.7,"publicationDate":"2022-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46926438","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A plague on both your houses: The debate about how to deal with ‘inconclusive’ conclusions when calculating error rates","authors":"G. Morrison","doi":"10.1093/lpr/mgac015","DOIUrl":"https://doi.org/10.1093/lpr/mgac015","url":null,"abstract":":","PeriodicalId":48724,"journal":{"name":"Law Probability & Risk","volume":null,"pages":null},"PeriodicalIF":0.7,"publicationDate":"2022-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41246369","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Erosion of the basic human rights","authors":"Nugzar Saria","doi":"10.52340/law.2022.01","DOIUrl":"https://doi.org/10.52340/law.2022.01","url":null,"abstract":"","PeriodicalId":48724,"journal":{"name":"Law Probability & Risk","volume":null,"pages":null},"PeriodicalIF":0.7,"publicationDate":"2022-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81777473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Inconclusives are not errors: a rejoinder to Dror","authors":"H. Arkes, J. Koehler","doi":"10.1093/lpr/mgac009","DOIUrl":"https://doi.org/10.1093/lpr/mgac009","url":null,"abstract":"","PeriodicalId":48724,"journal":{"name":"Law Probability & Risk","volume":null,"pages":null},"PeriodicalIF":0.7,"publicationDate":"2022-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49463500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As the basis of criminal penalty, criminal conviction, integral to the protection of fundamental rights and freedom of people constitutes the basis and the core issue of criminal trials. Based on the data published on China Judgments Online, we proposed two types of classification models to apply the data of four common crimes from China Judgments Online and expounded their applications in identifying ‘abnormal cases’, defined as wrongly sentenced cases in this article. The two types of classification models we proposed are a two-stage model and two deep learning models. To construct the two-stage model, we first used three keyword-extraction models to extract the keywords and vectorize all the keywords, then used five classification models to build the two-stage model. For the deep learning models, we applied two different deep neural network models in the data to build the classifier. We then applied these two types of classification models to discover ‘abnormal cases’ in two steps. In the first step, we applied the two-stage model to extract the ‘important words’ which will significantly improve the probability of the two-stage model to classify cases into crimes of intentional injury. In the second step, we constructed a validation data set of cases whose verdicts are changed in the second instance rulings to test the ‘important words’ extracted in first step and the ability of the two-stage model and the two deep learning models to discover ‘abnormal cases’. The results of this exercise show that: (1) ‘important words’ extracted in the first step are often associated with ‘abnormal cases’; (2) these two types of classification models can effectively discover ‘abnormal cases’, but compared with the two deep learning models, the two-stage model (aka. Term Frequency-Inverse Document Frequency and Artificial Neural Network, the combination of a keyword extraction model and a classic machine-learning model) is more capable of discovering ‘abnormal cases’.
{"title":"‘This Crime is Not That Crime’—Classification and evaluation of four common crimes","authors":"K. Xu, Han Liu, Fang Wang, Han Wang","doi":"10.1093/lpr/mgac006","DOIUrl":"https://doi.org/10.1093/lpr/mgac006","url":null,"abstract":"\u0000 As the basis of criminal penalty, criminal conviction, integral to the protection of fundamental rights and freedom of people constitutes the basis and the core issue of criminal trials. Based on the data published on China Judgments Online, we proposed two types of classification models to apply the data of four common crimes from China Judgments Online and expounded their applications in identifying ‘abnormal cases’, defined as wrongly sentenced cases in this article. The two types of classification models we proposed are a two-stage model and two deep learning models. To construct the two-stage model, we first used three keyword-extraction models to extract the keywords and vectorize all the keywords, then used five classification models to build the two-stage model. For the deep learning models, we applied two different deep neural network models in the data to build the classifier. We then applied these two types of classification models to discover ‘abnormal cases’ in two steps. In the first step, we applied the two-stage model to extract the ‘important words’ which will significantly improve the probability of the two-stage model to classify cases into crimes of intentional injury. In the second step, we constructed a validation data set of cases whose verdicts are changed in the second instance rulings to test the ‘important words’ extracted in first step and the ability of the two-stage model and the two deep learning models to discover ‘abnormal cases’. The results of this exercise show that: (1) ‘important words’ extracted in the first step are often associated with ‘abnormal cases’; (2) these two types of classification models can effectively discover ‘abnormal cases’, but compared with the two deep learning models, the two-stage model (aka. Term Frequency-Inverse Document Frequency and Artificial Neural Network, the combination of a keyword extraction model and a classic machine-learning model) is more capable of discovering ‘abnormal cases’.","PeriodicalId":48724,"journal":{"name":"Law Probability & Risk","volume":null,"pages":null},"PeriodicalIF":0.7,"publicationDate":"2022-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44673809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：