Evans, James and Pedro Aceves. 2016. “Machine Translation: Mining Text for Social Theory”. Annual Review of Sociology 42:21-50. DOI: 10.1146/annurev-soc-081715-074206
Thanks for the profound introduction to the history of computational content analysis and the possibility of the variety of information embedded in texts. However, I’m still a little bit confused about how to assess the reliability of findings via unsupervised methods and how to convince people that the machines have provided the correct interpretation of the content. Are there any statistical standards such as accuracy rate in supervised machine learning to validate the findings via unsupervised methods in content analysis besides its confirmation of theoretical significance?
Thanks for the profound introduction to the history of computational content analysis and the possibility of the variety of information embedded in texts. However, I’m still a little bit confused about how to assess the reliability of findings via unsupervised methods and how to convince people that the machines have provided the correct interpretation of the content. Are there any statistical standards such as accuracy rate in supervised machine learning to validate the findings via unsupervised methods in content analysis besides its confirmation of theoretical significance?