表象式语义网络研究

Method of Semantic Nets Based on Mental Imagery

摘要: 为解决形象思维及表象信息相关问题,结合语义网络以及表象式方法各自的优点,提出一种表象式语义网络知识表示方法.在语义网络中加入表象信息可以实现形象思维相关的推理.对表象式语义网络的构成要素及网络的组织进行了讨论,基于蚁群和最大最小蚁群算法提出了一种求解表象式语义网络的算法,给出了算法的实现步骤.实验结果表明,用表象式语义网络表示知识是可行的,利用该算法可以在表象式语义网络上进行推理,并能找到较为理想的解.

Abstract: The authors combine the advantages of semantic nets and mental imagery methodology, and present a method of relationship model representation for computer.That is semantic nets of mental imagery methodology.Joining the representation of information is in order to achieve related reasoning of imagery thinking.The object-oriented method describes the thing being studying doubly, viz.static describing and dynamic describing.The authors present a IAS algorithm based on AS and MMACS algorithm, and give the operation steps of IAS algorithm.The experimental results show that IAS algorithm is able to find a better solution for mental imagery semantic nets.