Social prediction: a new research paradigm based on machine learning

Abstract Sociology is a science concerned with both the interpretive understanding of social action and corresponding causal explanation, process, result. A explanation should be foundation prediction. For many years, due to data computing power constraints, quantitative research in has primarily focused on statistical tests analyze correlations causality, leaving predictions largely ignored. By sorting out historical context "social prediction," this article redefines concept by introducing why how machine learning can help prediction scientific way. Furthermore, summarizes academic value governance suggests that it potential breakthrough contemporary paradigm. We believe through learning, we witness advent an era paradigm shift from correlation causality This will provide rare opportunity for sociology China become international frontier computational sciences accelerate construction philosophy Chinese characteristics.