Analysis on Drug Dosage Form Name Based on N-gram Technique and Network Analysis

In this paper, we analyzed drug dosage form names. We created the network structure whose nodes are dosage form names. Its edges between dosage form names denote that they share some of sub-strings generated based on N-gram technique. We employed Simpson coefficient to define the weight of an edge. We proposed a new clustering method and applied it to the network. The results showed that “dosage forms” can be categorized based on not only physical form information but their application site,purpose,processing and so on.