Abstract
In dealing with the Semantic similarity of Chinese words, using Hownet would be a preferable popular method. But HowNet has its own shortcomings. First, most of the concepts in HowNet concern in explaining common terms, lacking of specialized vocabulary. Second, most of the concepts in HowNet are independent concepts and lack of compound concepts. In response to the above problems, we combine HowNet with Conceptual Knowledge Tree Knowledge Representation System. Based on this, we have analyzed the structures of concepts, distributed different weights to relevant positions, determined their meanings and calculated Chinese semantic similarity of automation discipline. The results are satisfactory basically.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
