用混合模型分析社交媒体帖子，自动识别自杀风险等级。
Detection of Suicidal Risk on Social Media: A Hybrid Model
- 融合RoBERTa与TF-IDF-PCA，结合深度语义与统计特征
- 在四分类任务中达0.7512的加权F1分数，优于单一模型
- 适合心理安全监测、早期干预系统研发者参考
自杀念头和行为日益被视为重要的社会问题，亟需有效的工具实现早期风险识别。本文构建了基于Reddit帖子的机器学习模型，将文本自动分类为四个自杀风险等级。采用多分类任务框架，提出一种融合罗伯特（RoBERTa）深度上下文嵌入、TF-IDF统计词权重及主成分分析（PCA）压缩的混合模型，以提升评估准确性和可靠性。针对数据不平衡与过拟合问题，探索多种重采样与数据增强策略以改善泛化能力。实验表明，该混合模型表现更优，最高加权F1得分为0.7512，显著优于仅使用RoBERTa、BERT及其他传统机器学习分类器的结果。
原文摘要 · Abstract (English)
Suicidal thoughts and behaviors are increasingly recognized as a critical societal concern, highlighting the urgent need for effective tools to enable early detection of suicidal risk. In this work, we develop robust machine learning models that leverage Reddit posts to automatically classify them into four distinct levels of suicide risk severity. We frame this as a multi-class classification task and propose a RoBERTa-TF-IDF-PCA Hybrid model, integrating the deep contextual embeddings from Robustly Optimized BERT Approach (RoBERTa), a state-of-the-art deep learning transformer model, with the statistical term-weighting of TF-IDF, further compressed with PCA, to boost the accuracy and reliability of suicide risk assessment. To address data imbalance and overfitting, we explore various data resampling techniques and data augmentation strategies to enhance model generalization. Additionally, we compare our model's performance against that of using RoBERTa only, the BERT model and other traditional machine learning classifiers. Experimental results demonstrate that the hybrid model can achieve improved performance, giving a best weighted $F_{1}$ score of 0.7512.
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。