CHiPS: Character Histograms and Positional Signals for Lightweight Authorship Attribution in Romanian Texts

CHiPS：用于罗马尼亚语文本轻量级作者归属分析的字符直方图与位置信号方法

Abstract: We propose CHiPS, a lightweight character-level authorship attribution method for Romanian texts. All reported experiments are closed-set: the true author is one of the candidate authors in the training data. CHiPS studies two complementary fingerprints of writing style: CH-SVM, a character-histogram classifier based on one-character marginal distributions, and FFT12-LR, a positional-signal classifier that represents selected characters and punctuation classes as impulse trains (binary indicator sequences over character positions) and extracts Fourier/Welch spectral descriptors.

摘要： 我们提出了 CHiPS，这是一种针对罗马尼亚语文本的轻量级字符级作者归属分析方法。所有报告的实验均为封闭集实验：即真实作者包含在训练数据的候选作者名单中。CHiPS 研究了两种互补的写作风格指纹：CH-SVM（一种基于单字符边际分布的字符直方图分类器）和 FFT12-LR（一种位置信号分类器，它将选定的字符和标点符号类别表示为脉冲序列，即字符位置上的二进制指示序列，并提取傅里叶/韦尔奇谱描述符）。

We also report CHiPS-F, a leakage-safe decision-level fusion variant, and an optional top-5 listwise reranker trained only on out-of-fold predictions. The method requires no tokenization, syntactic analysis, pretrained language model, or transformer fine-tuning, and it avoids character $n$-gram features with $n \geq 2$ in the histogram component.

我们还报告了 CHiPS-F，这是一种防泄漏的决策级融合变体，以及一个仅在折外预测（out-of-fold predictions）上训练的可选前 5 名列表重排序器。该方法不需要分词、句法分析、预训练语言模型或 Transformer 微调，并且在直方图组件中避免了 $n \geq 2$ 的字符 $n$-gram 特征。

On a locked grouped ROST split comprising 400 files from 392 source-text groups, written by 10 authors, with source-text-level evaluation and grouped five-fold model selection, CHiPS-F reaches 0.9310 accuracy and 0.9341 macro-F1. A matched but unrestricted character 2—5-gram TF—IDF SVM comparator reaches 1.0000 accuracy and macro-F1 on the same held-out groups, so the contribution is not a claim of best possible classification accuracy. Instead, the experiments ask how far restricted, transparent character evidence can go under strict leakage control.

在一个包含 392 个源文本组、共 400 个文件、由 10 位作者撰写的锁定分组 ROST 数据集上，通过源文本级评估和分组五折模型选择，CHiPS-F 达到了 0.9310 的准确率和 0.9341 的宏观 F1 分数。在相同的留出组上，一个匹配但无限制的字符 2-5-gram TF-IDF SVM 对照组达到了 1.0000 的准确率和宏观 F1 分数，因此本研究的贡献并非声称达到了最佳的分类准确率。相反，这些实验旨在探讨在严格的泄漏控制下，受限且透明的字符证据究竟能达到何种效果。

On ROSTories-cleaned, a secondary ROST-overlapping corpus comprising 1,248 files from 1,240 source-text groups, written by 19 authors, the same protocol gives 0.8919 accuracy and 0.8708 macro-F1 for CHiPS-R.

在 ROSTories-cleaned（一个与 ROST 重叠的二级语料库，包含 1,240 个源文本组、共 1,248 个文件，由 19 位作者撰写）上，采用相同的协议，CHiPS-R 获得了 0.8919 的准确率和 0.8708 的宏观 F1 分数。