Prediction of highly cited papers

摘要

In an article written five years ago [arXiv:0809.0522], we described a methodfor predicting which scientific papers will be highly cited in the future, evenif they are currently not highly cited. Applying the method to real citationdata we made predictions about papers we believed would end up being wellcited. Here we revisit those predictions, five years on, to see how well wedid. Among the over 2000 papers in our original data set, we examine the fiftythat, by the measures of our previous study, were predicted to do best and wefind that they have indeed received substantially more citations in theintervening years than other papers, even after controlling for the number ofprior citations. On average these top fifty papers have received 23 times asmany citations in the last five years as the average paper in the data set as awhole, and 15 times as many as the average paper in a randomly drawn controlgroup that started out with the same number of citations. Applying ourprediction technique to current data, we also make new predictions of papersthat we believe will be well cited in the next few years.