Unsupervised prediction of citation influences 论文

摘要

Abstract Publication repositories contain an abundance of information about the \nevolution of scientific research areas. We address the problem of creating a \nvisualization of a research area that describes the flow of topics between \npapers, quantifies the impact that papers have on each other, and helps to \nidentify key contributions. To this end, we devise a probabilistic topic model \nthat explains the generation of documents; the model incorporates the aspects \nof topical innovation and topical inheritance via citations. We evaluate the \nmodel's ability to predict the strength of influence of citations against \nmanually rated citations.