arXiv:2508.18208cs.CL2025-08被引 1

通过语言分析,发现不同音乐粉丝在性格上存在显著差异。

On the Interplay between Musical Preferences and Personality through the Lens of Language

  • 用5000多名作者的文本,分析音乐偏好与性格的关系。
  • 发现五种音乐类型听众在五大性格维度上表现明显不同。
  • 适合对心理分析、语言建模和音乐认知感兴趣的学者。

音乐是个人身份的重要反映,常与深层心理特质相关。已有研究证实音乐偏好与人格之间存在关联,而另一些研究则表明人格可通过语言分析识别。本研究将这两个领域结合,从大五人格(开放性、尽责性、外向性、宜人性、神经质)视角出发,探究个体音乐偏好是否会在其自发语言中留下痕迹。基于包含近5000名作者、超过50万条文本样本的精心构建数据集,我们建立了先进模型以评估人格特征。结果显示,不同音乐类型听众在五大人格维度上存在显著差异。研究资源已公开,供计算语言学、音乐心理学与人格分析交叉领域的后续研究使用。

原文摘要 · Abstract (English)

Music serves as a powerful reflection of individual identity, often aligning with deeper psychological traits. Prior research has established correlations between musical preferences and personality, while separate studies have demonstrated that personality is detectable through linguistic analysis. Our study bridges these two research domains by investigating whether individuals' musical preferences leave traces in their spontaneous language through the lens of the Big Five personality traits (Openness, Conscientiousness, Extroversion, Agreeableness, and Neuroticism). Using a carefully curated dataset of over 500,000 text samples from nearly 5,000 authors with reliably identified musical preferences, we build advanced models to assess personality characteristics. Our results reveal significant personality differences across fans of five musical genres. We release resources for future research at the intersection of computational linguistics, music psychology and personality analysis.

人格分析语言模型音乐心理

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。