Egami 等人的工作通过展示如何从文本数据(其中文本可以是结果或干预)中量化估计因果关系,推进文本分析和因果推断这一交叉学科。他们的论文以及越来越多关于从文本中进行因果推理的文献中发展的方法,可以帮助社会科学家和其他研究人员从文本数据中进行新类型的因果推理。参考文献1 J. Pearl, Causality (Cambridge Univ. Press, ed. 2, 2009).2 G. Imbens, D. Rubin, Causal Inference in Statistics, Social and Biomedical Sciences: An Introduction (Cambridge Univ. Press, 2015).3 S. Morgan, C. Winship, Counterfactuals and Causal Inference (Cambridge University Press, ed. 2, 2015).4 M. Hernan, J. Robins, Causal Inference: What If? (Chapman & Hall/CRC, 2020).5 N. Egami, C. Fong, J. Grimmer, M. Roberts, B. Stewart, How to make causal inferences using texts. Sci. Adv. 8, eabg2652 (2022).6 V. Veitch, D. Sridhar, D. Blei, Uncertainty in Artificial Intelligence (Proceedings of Machine Learning Research, 2020).7 M. E. Roberts, B. M. Stewart, D. Tingley, C. Lucas, J. Leder-Luis, S. K. Gadarian, B. Albertson, D. G. Rand, Structural topic models for open-ended survey responses. Am. J. Polit. Sci. 58, 1064–1082 (2014).8 O. Netzer, A. Lemaire, M. Herzenstein, When words sweat: Identifying signals for loan default in the text of loan applications. J. Market. Res. 56, 960–980 (2019).9 C. Fong, J. Grimmer, Causal inference with latent treatments. Am. J. Polit. Sci. 10.1111/ajps.12649 (2022).10 B. Schölkopf, F. Locatello, S. Bauer, N. Ke, N. Kalchbrenner, A. Goyal, Y. Bengio, Towards causal representation learning. arXiv:2102.11107 (2021).
本文翻译自 Science Advances 评论文章。原文题目:Causal inference from text: A commentary原文链接:https://www.science.org/doi/10.1126/sciadv.ade6585