Abstract
Understanding the fundamental concepts and trends in a scientific field iscrucial for keeping abreast of its continuous advancement. In this study, wepropose a systematic framework for analyzing the evolution of research topicsin a scientific field using causal discovery and inference techniques. Wedefine three variables to encompass diverse facets of the evolution of researchtopics within NLP and utilize a causal discovery algorithm to unveil the causalconnections among these variables using observational data. Subsequently, weleverage this structure to measure the intensity of these relationships. Byconducting extensive experiments on the ACL Anthology corpus, we demonstratethat our framework effectively uncovers evolutionary trends and the underlyingcauses for a wide range of NLP research topics. Specifically, we show thattasks and methods are primary drivers of research in NLP, with datasetsfollowing, while metrics have minimal impact.