A context dependency relation pattern based algorithm for term similarity calculation

Abstract

In this paper, we put forward a new term similarity algorithm which automatically extracts context patterns by sentence dependency analysis, and then calculates term similarity by mapping those patterns. The algorithm provides better way to extract term context patterns with stronger context expression ability than current algorithms. Meanwhile, the length of patterns is modest, which makes it convenient to be matched. We also evaluate the algorithm on the basis of gene engineering field experiment data set. Experiment result shows that the algorithm evidently improves the calculating performance.