Speech and Language Laboratory, the Australian National University, Canberra, Australia; Linguistics Program, School of Culture, History and Language, College of Asia and the Pacific, the Australian National University, Building #110, Canberra, ACT 2600, Australia.
Forensic Sci Int. 2022 May;334:111268. doi: 10.1016/j.forsciint.2022.111268. Epub 2022 Mar 10.
This study compares score- and feature-based methods for estimating forensic likelihood ratios for text evidence. Three feature-based methods built on different Poisson-based models with logistic regression fusion are introduced and evaluated: a one-level Poisson model, a one-level zero-inflated Poisson model and a two-level Poisson-gamma model. These are compared with a score-based method that employs the cosine distance as a score-generating function. The two types of methods are compared using the same data (i.e., documents attributable to 2,157 authors) and the same features set, which is a bag-of-words model using the 400 most frequently occurring words. Their performances are evaluated via the log-likelihood ratio cost (C) and its composites: discrimination (C) and calibration (C) cost. The results show that (1) the feature-based methods outperform the score-based method by a C value of 0.14-0.2 when their best results are compared and (2) a feature selection procedure can further improve performance for the feature-based methods. Some distinctive performance characteristics associated with likelihood ratios produced using the feature-based methods are described, and their implications will be discussed with real forensic casework in mind.
本研究比较了基于评分和特征的方法，以估计文本证据的法医似然比。介绍并评估了三种基于特征的方法，这些方法基于不同的泊松模型，并结合逻辑回归融合：一级泊松模型、一级零膨胀泊松模型和两级泊松-伽马模型。这些方法与基于评分的方法进行了比较，该方法使用余弦距离作为评分生成函数。这两种方法使用相同的数据（即归因于 2157 位作者的文档）和相同的特征集进行比较，特征集是使用最常出现的 400 个单词的词袋模型。通过对数似然比成本（C）及其组合：区分（C）和校准（C）成本来评估它们的性能。结果表明：（1）当比较最佳结果时，基于特征的方法比基于评分的方法的 C 值高出 0.14-0.2；（2）特征选择过程可以进一步提高基于特征的方法的性能。描述了与使用基于特征的方法生成的似然比相关的一些独特性能特征，并将考虑实际法医案例讨论其含义。
