Title :
Text similarity calculation method based on ontology model
Author :
Tao Chi ; Hanshi Wang ; Lizhen Liu ; Wei Song ; Chao Du
Author_Institution :
Inf. & Eng. Coll., Capital Normal Univ., Beijing, China
Abstract :
Chinese text similarity calculation plays an important role in the Chinese information processing field. This paper uses ontology to calculate text similarity. The ontology has been widely used in natural language processing. Its unique concept model on the relationship offers a new processing method for word similarity calculation. Based on the characteristics of ontology and combined with HowNet and TongYiCi CiLin, we design a Hybrid Word Similarity Calculation-HWSC method and present a text similarity calculation method based on ontology model. The experimental results show that the method can make full use of the characteristics of ontology to accurately calculate the similarity between two texts.
Keywords :
natural language processing; ontologies (artificial intelligence); text analysis; word processing; Chinese information processing field; Chinese text similarity calculation method; HWSC method; HowNet; TongYiCi CiLin; hybrid word similarity calculation; natural language processing; ontology characteristics; ontology model; word similarity calculation; Ontologies; Transforms; HowNet; Ontology; Text similarity; TongYiCi CiLin;
Conference_Titel :
Cloud Computing and Internet of Things (CCIOT), 2014 International Conference on
Print_ISBN :
978-1-4799-4765-2
DOI :
10.1109/CCIOT.2014.7062539
