Self-awareness Evaluation Of Hearing Impairment Students Based On Bayesian Network

The hearing impairment people are a special group indispensable to the society.According to the latest statistics,the number of hearing impairment people has reached 20.75 million in China.Although they live in different parts of the country like healthy people,they have their own culture consciousness.Because there is no language stimulation,they have great difficulties in cognitive development and social adaptation.Because the hearing impairment people have low self-awareness and are unwilling to communicate with healthy people,the environment of their employment is directly affected and the quality of their lives is poor.Some hearing impairment people unable to integrate into the big family of society,so they go to the road of illegal crime.How to improve the level of self-awareness of the hearing impairment people,promote hearing impairment people integrate into society,improve the quality of their life,and reduce the number of crimes is an important issue for special education workers.Therefore,the research on the consciousness of the hearing impairment person is extra important.However,obtaining data from the hearing impairment person is difficult,and evaluation tool for hearing impairment person is simplification.Currently,this is only one tool to evaluate hearing impairment person's self-awareness which is the self-awareness measurement table.Self-awareness evaluation becomes very difficult,and the accuracy is often determined by factors such as the language comprehension and attitude of the man.Based on the above reasons,the thesis intends to build a self-awareness evaluation model for hearing impairment students based on Bayesian Network(BN,aiming to propose a new method to study the self-awareness of hearing impairment students.According to the model,special education workers will know self-awareness of hearing impairment students after understanding the relevant factors of hearing impairment students,and understand the relationship between the factors affecting the level of self-awareness of hearing impairment students,so as to help the hearing impairment students improving the level of self-awareness and providing better educational services for hearing impairment students.At first,the thesis finds out the factors related to the self-consciousness of the hearing impairment person,and then combines the prior knowledge of special educators to construct the self-consciousness evaluation BN model.Based on the initial model,we use mutual information to judge the dependence between any two random variables and then modify the initial BN model.since the data of the hearing impairment person is very small,and the data have missing items.Therefore,we use the Expectation Maximization(EM)algorithm to learn the Conditional Probability Table(CPT)of each node in the BN model.Based on the constructed BN,we use the variable elimination method to infer the level of self-consciousness.Evaluate the level of self-consciousness by giving the state of each factor,and according to the level of self-consciousness diagnose the influence of various factors.