Tracing the Historical Trajectory of Scientific Integration:AIGC Empowers Interdisciplinarity Measurement Research
[Purpose/Significancel The generative large language model has changed the paradigm of natural lan-guage processing research,promoted the new trend of AI-enabled social science research,and provided new ideas for quantitatively calculating disciplinary crossover and integration of humanities and social science disciplines from the per-spective of deep semantic features of texts.[Method/Process]In this paper,it used ChatGPT to discriminate the academic literature in humanities and social sciences.Based on few-shot learning,it recognized the discipline in the model predic-tion,measured the multidisciplinary classification from the perspective of the knowledge decentralized distribution,made a comparative analysis between the results and the corresponding disciplines of the journals.Finally,it put forward the indexes of cross-disciplinary richness,cross-disciplinary closeness,and subjectivity,combined with the interdisciplinary degree,interdisciplinary quantitative research.[Result/Conclusion]This paper focuses on AIGC-enabled interdisciplin-arity measurement,proposes a set of research framework methods from the judgment of disciplinary attribution,the ex-traction of disciplinary names in the answer set of the generative model,the empowerment of multidisciplinary candidate problems,and interdisciplinarity content-based metric indicators,and realizes the AIGC-enabled social sciences research from the content.It provides a reference for the internal logic of social science research.
ChatGPTlarge language modelinterdisciplinarity measurementlaw of Bradfordhuman-ities & social sciencestime series forecasting analysis
万方数据
维普
