Research On Associative Knowledge Network Modeling And Its Application

Text semantic representation is one of the core contents of natural language processing and one of the most basic tasks in text understanding,processing and searching.Traditional text representation methods are often unable to adequately model the contextual information of the text,which makes the structural information of the text missing and the captured semantic information insufficient.The method based on neural network can get relatively good results,but its poor interpretability seriously affects its application range.Considering that knowledge usually exists in the form of associative memory in human brain,this paper explores the internal structure of knowledge system from the perspective of human brain associative memory,in order to explore a new text semantic modelling technology,and carries out corresponding application research.Uninterpretability has become the biggest obstacle to the wider application of deep neural network,especially in most human-machine interaction scenes.Inspired by the powerful associative computing ability of human brain neural system,a novel interpretable semantic representation model of noun context,associative knowledge network model,is proposed.The proposed network structure is composed of only pure associative relationships without relation label and is dynamically generated by analysing neighbour relationships between noun words in text,in which incremental updating and reduction reconstruction strategies can be naturally introduced.Furthermore,aiming at the intelligent error correction application scenario of text writing,a novel interpretable method is designed for the practical problem of checking the semantic coherence of noun context.In proposed method,the associative knowledge network learned from the text corpus is first regarded as a background knowledge network,and then the multilevel contextual associative coupling degree features of noun words in given detection document are computed.Finally,contextual coherence detection and the location of those inconsistent noun words can be realized by using an interpretable classification method such as decision tree.Our sufficient experimental results show that above proposed method can obtain excellent performance and completely reach or even partially exceed the performance obtained by the latest deep neural network methods.In addition,starting from the task of Chinese part-of-speech tagging,this paper proposes a Chinese part-of-speech tagging method based on associative knowledge network.The new method firstly constructs an associative knowledge network with attributes.Next,the sentences that need part-of-speech tagging are put into the associative knowledge network for walking analysis to generate multiple walking pathways.Then,the concept of pathway evolution intensity is introduced to calculate the strengths of the walking pathways,so as to get optimal pathway.Finally,the part-of-speech tagging results of sentences are obtained according to the nodes and attributes on the optimal pathway.Experimental analysis and comparison results show the effectiveness of the new method,and provide a new idea for Chinese part-of-speech tagging.The method proposed in this paper has the incomparable advantages over neural network in terms of natural interpretability and incremental learning ability.It provides a very enlightening idea for developing interpretable machine learning methods,especially in text semantic representation modelling and its practical application.