Studying The Construction Of Chinese Surface Semantic Neural Network Based On Ontology

Semantic Neural Network is a new methodology for Natural Language Understanding that combines symbolism and connection- ism. It is considered that the Natural Language Understanding is the process that the linguistic labels activate the corresponding neuron in the human brain, and set up or activate the Semantic Neural Network. But this research is just at the beginning, researchers only provide the idea and its models. In the other hand, Chinese understanding is always difficult. No matter regular method or statistic method, none of them systematically proposes an analytical method according to the characteristics of Chinese. We find that the Semantic Neural Network relies on semantic analysis mainly, which just accords with the characteristics of Chinese as the language of analytical type. Meanwhile, the neuron is connected according to semantic relation between the concept, which is very good to solve the difficult problems that the fuzzy on semantic and the differentiation of morphology. The paper is come to simulate and realize tentatively the model of Semantic Neural Network, and combine with the semantic characteristic of Chinese to realize the understanding of the surface semantics of Chinese.In this paper, we design the model of the neuron in the Semantic Neural Network,present the description of the attribute and method, and the definition of the influences. The Conceptual Ontology, based on the analysis of the Chinese Surface Semantics relation, is also constructed. We present the definition and the description of Conceptual Ontology, and set up the relation of mapping between it and neurons. On the basis of this, the mechanism of negotiation to guarantee the orderly connection between the neurons that can act spontaneously is introduced, the negotiation model and negotiation languages are defined, and the negotiation algorithm is also presented. Thereby, the Natural Language Understanding is the process that the corresponding neurons binding correlative Concept Ontology knowledge manage to agreement via mutual operation and negotiation. At last through the analysis of some examples, it is apparent that the feasibility of Semantic Neural Network.