Entity Concept Knowledge Database Construction Oriented To Metaphor Computation

Besides researching on computational mechanisms of metaphor within the last few decades, great attention has also been paid to knowledge database construction which can be applied into the computational model of metaphor interpretation. Knowledge is the basis for metaphor understood by machine. Take "Motherland is like mother" for example, machine can only know the mean of this sentence on the base of achieving the knowledge of the words "motherland" and "mother" from the knowledge database.This dissertation addresses the research on the knowledge database construction for metaphor interpretation. We firstly proposed a method of entity concept knowledge database construction and finally got it with a certain scale. Then we applied the knowledge database into the Chinese metaphor computational model and got a good result.Our research includes the following aspects:1. Introduced the background and significance of our research. And made a summarization of exiting research in this field.2. Made the semantic analysis and introduced the definition and features of entity knowledge.3. Proposed and carried out the way of entity knowledge database construction. Firstly, we generated entity concept database mainly basing on hownet, and attribute value database using statical way. Further more, we defined the measure and computation method of attribute determining, and creatively brought forward the computation model of ambiguous attribute determination——semantic and example based. Finally a full database was built and the experiment result was analyzed and showed good.4. Proposed a new computational model and apply entity concept database into this new metaphor computational model.Our research of entity concept knowledge database construction used for Chinese Metaphor computation is a good innovation, and according to the experimental results, it shows that the method is feasible.