开源工具PUCP-Metrix提供182项西班牙语语言分析指标，助力文本可解释性研究。
PUCP-Metrix: An Open-source and Comprehensive Toolkit for Linguistic Analysis of Spanish Texts
- 整合182项语言特征，覆盖词汇多样性到可读性等多个维度。
- 在可读性评估与机器生成文本检测任务中表现媲美现有基准。
- 支持多场景自然语言处理应用，适合研究者与开发者使用。
语言特征对风格、结构和可读性等任务的可解释性至关重要，但现有西班牙语工具覆盖有限。本文提出PUCP-Metrix，一个开源且全面的西班牙语文本语言分析工具包。该工具包包含182项语言度量，涵盖词汇多样性、句法与语义复杂性、衔接性、心理语言学及可读性等方面，支持细粒度、可解释的文本分析。我们在自动可读性评估和机器生成文本检测任务上验证其有效性，结果表明其性能与现有资源及强基线神经模型相当。PUCP-Metrix为西班牙语提供了全面且可扩展的资源，支持多样化的自然语言处理应用。
原文摘要 · Abstract (English)
Linguistic features remain essential for interpretability and tasks that involve style, structure, and readability, but existing Spanish tools offer limited coverage. We present PUCP-Metrix, an open-source and comprehensive toolkit for linguistic analysis of Spanish texts. PUCP-Metrix includes 182 linguistic metrics spanning lexical diversity, syntactic and semantic complexity, cohesion, psycholinguistics, and readability. It enables fine-grained, interpretable text analysis. We evaluate its usefulness on Automated Readability Assessment and Machine-Generated Text Detection, showing competitive performance compared to an existing repository and strong neural baselines. PUCP-Metrix offers a comprehensive and extensible resource for Spanish, supporting diverse NLP applications.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。