Study On Application Of Evaluation Model For Primary And Middle School Student Education

Cluster analysis is a kind of clustering algorithm according to the data implied category information for unsupervised.In this paper,On the one hand,clustering center of quantitative attribute variable is calculated by means.On the other hand,clustering center of qualitative attribute variable is calculated by two ways.One way is calculated by mode,the other way which the fuzzy clustering center of qualitative attribute variables is calculated by the frequency value of each attribute value of each attribute is based on the way.In terms of similarity measure,in the first way,quantitative attribute variable is calculated by Euclidean distance while qualitative attribute variable is calculated by 0-1 matching and let’s sum over two distances.Then the formula need a parameter which obtained according to the actual situation to weigh the distance between the quantitative attribute variable and the qualitative attribute variable.According the data information,the final value of is 0.5.In the second way,based on the fuzzy clustering center of qualitative attribute variables,all attributes are measured by Euclidean distance.Finally,let’s sum over two distances as the distance between the samples.Based on the second way,the weight of the objective function is based on the information entropy.In the objective function weight calculation,the distribution of each attribute is considered.Based on three kinds of clustering algorithms are improved in this paper,the data of primary and middle school students are analyzed.From the results,we can observe that the hybrid data clustering analysis algorithm based on information entropy has the minimum of the sum of squares within a class which has the best clustering effect.Aim at the typical cases of students which are finally determined by hybrid data clustering analysis based on information entropy,eight rotation factors extracted from the factor analysis model based on the original data can clearly explain the information implied by the original 55 quantitative attribute variables.In combination with the coefficients of factor value obtained through factor analysis,the factor scores of typical cases of students in each class obtained through the hybrid data clustering analysis algorithm based on information entropy are calculated.The evaluation system was established by using the clustering center of qualitative attribute variables and the value of factor scores of the clustering center of quantitative attribute variable.From the system,we can find Difficulties and problems coming from students in study and life,and solve the questions and give reasonable opinions and suggestions of typical cases of students in each class.