Research On The Application Of Nmli Model In Cognitive Behavior Analysis And Pattern Recognition

The NMLI model is a new type of neural model,which is built by simulating the activity patterns and connectivity characteristics of human brain neurons at different levels.It has good learning performance in small samples,possesses good potential for application in the field of cognition and machine learning,and has produced certain positive results in image classification and clustering problems.However,there are some shortcomings in the current research on this model.For example,the dynamics of the model has not been studied in depth,and the existence and stability of the solution of the model have been rarely discussed;the incompatibility of existing methods with human cognitive characteristics,when dealing with image classification tasks,limits use of this model for machine learning problems such as image classification.Therefore,this paper addresses these two problems of the NMLI model,and the main work is as follows:First,this paper uses the NMLI model to describe the acquisition and consolidation processes of the fear memory model in Obsessive-Compulsive Disorder(OCD).The existence and stability of the static solution of the model are investigated,and the conditions for the acquisition and consolidation of the fear memory model are obtained.This paper also gives a possible explanation for the occurrence of specific cognitive phenomena such as OCD and opens the way for further applications of the NMLI model to cognitive problems.Secondly,by modelling the concentration phenomenon in visual cognition,a new concentration mechanism based on information entropy is developed for the NMLI model.With this mechanism,the model focuses its attention on the parts of the image data that are closely related to classification,and optimizes the structure of the NMLI model to effectively improve the performance of the model in image recognition tasks.Numerical experiments show that the computational efficiency and classification accuracy of the improved NMLI model with the concentration mechanism are effectively improved compared with the original model,which improves the applicability of the NMLI model in the engineering field.