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
According to the reading, the computational approach enables sociologists to do large-scale analysis of the text by translating them into sociologically relevant data. However, the meaning of a text is not always stable over time. A certain word or phrase can have different cultural symbolic meaning in different time. How can text analysis overcome this problem? (e.g. In Chinese, "river crab" also used to refer to a "harmonious society" in recent years because they sound like each other. It's a kind of evasion from the censorship.)
According to the reading, the computational approach enables sociologists to do large-scale analysis of the text by translating them into sociologically relevant data. However, the meaning of a text is not always stable over time. A certain word or phrase can have different cultural symbolic meaning in different time. How can text analysis overcome this problem? (e.g. In Chinese, "river crab" also used to refer to a "harmonious society" in recent years because they sound like each other. It's a kind of evasion from the censorship.)