IIRank: co-ranking scientific articles by characterizing their innovativeness and importance

The dynamic nature of literature networks makes the task of ranking scientific articles hard, hence we present a framework(IIRank) to determine the influence of the proportion of feature pairs in the scientific article on the innovation of the article and to co-rank the scientific article. The model is based on the title, keyword and abstract information extracted from the scientific article, which make it possible to consider the feature pair of the scientific article as the sensors of their innovativeness and importance, and we use the Entropy method to judge the importance degree of a feature pair, the greater the importance degree, the greater the impact on the comprehensive evaluation.