A RMM Based Word Segmentation Method for Chinese Design Specifications of Building Stairs

Abstract

With the rapid development of information technology, knowledge graph extracts more and more attentions from researchers. However, Chinese knowledge graph of construction industry is still at the beginning stage, and Chinese word segmentation method, as the basis of natural language processing, plays a vital role on the process of building knowledge graph. In this paper, we study Chinese design specifications of building stairs, and proposes a reverse maximum matching (RMM) based word segmentation method to parse Chinese building specifications. The proposed method first converts the dictionary of building into a hash dictionary. And then, by traversing the design specifications of building stairs, the proposed method handles the non-Chinese symbols in the design specifications. Finally, the proposed method uses RMM algorithm to match contexts with Chinese design specifications and generate the goal results. Through performing experiments on Chinese design specifications of building stairs, the results can be shown that the proposed method is feasible.