Using Citation-KNN for Automatic Keyword Assignment

Currently, the automatic keywords extraction method can only extract keywords appeared in the articles and it cannot extract the implicit keyword which does not appear in the articles. It is a difficult work to extract implicit keywords in an article in the task of automatic keywords extraction. This work can also be called automatic keyword assignment. In this paper, an automatic keyword assignment method based on Citation-KNN (Citation-K Nearest Neighborhood) is proposed. Experimental results show that the proposed method can not only improve the precision and recall of keyword extraction, but also extract implicit keyword which does not appear in the articles efficiently.