基于持续关注度衰减的重要论文预测

Citation Prediction Based on Sustained Attention Decay

针对现有方法存在预测精度较低或者计算复杂等问题,通过真实数据集的实证分析,发现论文的年均引用次数与论文未来的被引次数有很大的相关性,由此提出持续关注度的概念。进一步,结合论文引用的时间衰减特性,提出一种基于持续关注度衰减的重要论文预测算法。在两个典型数据集上的实验结果表明,该方法不仅计算简单,而且具有较高的预测精度。

Existing methods have problems such as low accuracy or time consuming. Through em- pirical analysis of real data sets, we find that there is a strong correlation between future citations and annual citation of papers, i.e. the sustained attention; meanwhile, the future citations have apparent characteristics of time decay. Thus we put forward a method based on sustained atten- tion decay to predict the future citations of papers, and then to find papers of potential importance with these results. Experimental results on two benchmark data sets show that the proposed method can predict precisely with lower time complexity.