人工智能在911凶杀电话中检测骗局的工具

IF 3.3 1区社会学 Q1 CRIMINOLOGY & PENOLOGY

Journal of Criminal Justice Pub Date : 2025-01-01 DOI:10.1016/j.jcrimjus.2024.102337

Patrick M. Markey, Samantha Goldman, Jennie Dapice, Sofia Saj, Saadet Ceynek, Tia Nicolas, Lila Trollip

{"title":"人工智能在911凶杀电话中检测骗局的工具","authors":"Patrick M. Markey, Samantha Goldman, Jennie Dapice, Sofia Saj, Saadet Ceynek, Tia Nicolas, Lila Trollip","doi":"10.1016/j.jcrimjus.2024.102337","DOIUrl":null,"url":null,"abstract":"<div><div>This paper investigates the application of Artificial Intelligence (AI), specifically the use of a Large Language Model (ChatGPT), in analyzing 911 calls to identify deceptive reports of homicides. The study sampled an equal number of False Allegation Callers (FACs) and True Report Callers (TRCs), categorized through judicial outcomes. Calls were processed using ChatGPT, which assessed 86 behavioral cues from 142 callers. Using a random forest model with k-fold cross-validation and repeated sampling, the analysis achieved an accuracy rate of 70.68 %, with sensitivity and specificity rates at 71.44 % and 69.92 %, respectively. The study revealed distinct behavioral patterns that differentiate FACs and TRCs. AI characterized FACs as somewhat unhelpful and emotional, displaying behaviors such as awkwardness, unintelligibility, moodiness, uncertainty, making situations more complicated, expressing regret, and self-dramatizing. In contrast, AI identified TRCs as helpful and composed, marked by responsiveness, cooperativeness, a focus on relevant issues, consistency, plausibility in their messages, and candidness.</div></div>","PeriodicalId":48272,"journal":{"name":"Journal of Criminal Justice","volume":"96 ","pages":"Article 102337"},"PeriodicalIF":3.3000,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Artificial intelligence as a tool for detecting deception in 911 homicide calls\",\"authors\":\"Patrick M. Markey, Samantha Goldman, Jennie Dapice, Sofia Saj, Saadet Ceynek, Tia Nicolas, Lila Trollip\",\"doi\":\"10.1016/j.jcrimjus.2024.102337\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>This paper investigates the application of Artificial Intelligence (AI), specifically the use of a Large Language Model (ChatGPT), in analyzing 911 calls to identify deceptive reports of homicides. The study sampled an equal number of False Allegation Callers (FACs) and True Report Callers (TRCs), categorized through judicial outcomes. Calls were processed using ChatGPT, which assessed 86 behavioral cues from 142 callers. Using a random forest model with k-fold cross-validation and repeated sampling, the analysis achieved an accuracy rate of 70.68 %, with sensitivity and specificity rates at 71.44 % and 69.92 %, respectively. The study revealed distinct behavioral patterns that differentiate FACs and TRCs. AI characterized FACs as somewhat unhelpful and emotional, displaying behaviors such as awkwardness, unintelligibility, moodiness, uncertainty, making situations more complicated, expressing regret, and self-dramatizing. In contrast, AI identified TRCs as helpful and composed, marked by responsiveness, cooperativeness, a focus on relevant issues, consistency, plausibility in their messages, and candidness.</div></div>\",\"PeriodicalId\":48272,\"journal\":{\"name\":\"Journal of Criminal Justice\",\"volume\":\"96 \",\"pages\":\"Article 102337\"},\"PeriodicalIF\":3.3000,\"publicationDate\":\"2025-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Criminal Justice\",\"FirstCategoryId\":\"90\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0047235224001867\",\"RegionNum\":1,\"RegionCategory\":\"社会学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"CRIMINOLOGY & PENOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Criminal Justice","FirstCategoryId":"90","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0047235224001867","RegionNum":1,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CRIMINOLOGY & PENOLOGY","Score":null,"Total":0}

引用次数: 0

摘要

本文研究了人工智能（AI）的应用，特别是使用大型语言模型（ChatGPT）来分析911电话，以识别欺骗性的凶杀案报告。该研究抽样了相同数量的虚假指控呼叫者（FACs）和真实报告呼叫者（TRCs），根据司法结果进行分类。电话使用ChatGPT处理，它评估了来自142个呼叫者的86个行为线索。采用k-fold交叉验证和重复采样的随机森林模型，分析准确率为70.68%，灵敏度和特异性分别为71.44%和69.92%。该研究揭示了区分FACs和TRCs的不同行为模式。AI将FACs描述为有些无助和情绪化，表现出尴尬、难以理解、喜怒无常、不确定、使情况变得更复杂、表达遗憾和自我戏剧化等行为。相比之下，人工智能认为trc是有帮助的和组成的，其特点是响应性、合作性、对相关问题的关注、一致性、信息的合理性和坦率。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Artificial intelligence as a tool for detecting deception in 911 homicide calls

This paper investigates the application of Artificial Intelligence (AI), specifically the use of a Large Language Model (ChatGPT), in analyzing 911 calls to identify deceptive reports of homicides. The study sampled an equal number of False Allegation Callers (FACs) and True Report Callers (TRCs), categorized through judicial outcomes. Calls were processed using ChatGPT, which assessed 86 behavioral cues from 142 callers. Using a random forest model with k-fold cross-validation and repeated sampling, the analysis achieved an accuracy rate of 70.68 %, with sensitivity and specificity rates at 71.44 % and 69.92 %, respectively. The study revealed distinct behavioral patterns that differentiate FACs and TRCs. AI characterized FACs as somewhat unhelpful and emotional, displaying behaviors such as awkwardness, unintelligibility, moodiness, uncertainty, making situations more complicated, expressing regret, and self-dramatizing. In contrast, AI identified TRCs as helpful and composed, marked by responsiveness, cooperativeness, a focus on relevant issues, consistency, plausibility in their messages, and candidness.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Journal of Criminal Justice CRIMINOLOGY & PENOLOGY-

CiteScore

6.90

自引率

9.10%

发文量

审稿时长

23 days

期刊介绍： The Journal of Criminal Justice is an international journal intended to fill the present need for the dissemination of new information, ideas and methods, to both practitioners and academicians in the criminal justice area. The Journal is concerned with all aspects of the criminal justice system in terms of their relationships to each other. Although materials are presented relating to crime and the individual elements of the criminal justice system, the emphasis of the Journal is to tie together the functioning of these elements and to illustrate the effects of their interactions. Articles that reflect the application of new disciplines or analytical methodologies to the problems of criminal justice are of special interest. Since the purpose of the Journal is to provide a forum for the dissemination of new ideas, new information, and the application of new methods to the problems and functions of the criminal justice system, the Journal emphasizes innovation and creative thought of the highest quality.