Fuzzy predicting new association rules from current scientific literature

Paradoxically, the explosion of scientific information has resulted in diminishing awareness. In the face of an ever growing body of literature, disciplines are becoming increasingly specialized, while individuals and groups are becoming ever more insular. The availability of scientific bibliographies in online databases is a rich source of scientific information for scientists to support their research. In this paper, we propose a new method to predict new association rules of concepts by mining current scientific literature. In contrast to previous related research, our method´s novelties are as follows: extend the antecedent and consequent of an association rule from a concept to a set of concepts; measure the relationship between two concepts not only by their co-occurrence in scientific literature, but also by their inherent relationship in knowledge bases; describe the appropriate degree of replacing a concept with its sibling; propose some indicators to distinguish various valid changes of existing association rules. The predicted new association rules can serve researchers as major repositories of candidates for new research themes, as impetus for inspiration impetus, or as hypotheses to be tested in future.