Design And Implementation Of Teaching Management System Based On Data Mining In Vocational Colleges

With the large-scale enrollment expansion of Vocational Colleges in China,the number of students in various vocational colleges is increasing,among which the situation of students is becoming more and more complex.At present,the main direction of school information construction is data sharing and data integration analysis of different information systems.Educational administration system can share and analyze data with almost all relevant systems of Vocational colleges.Educational administration management system covers a large number of data of students in school,such as teaching effect data,morning service card data,achievement data,book borrowing and so on.By mining these data closely related to students' learning and life,the key information hidden behind the data is analyzed,and the instruction of teaching and infrastructure construction is obtained.Meaning has become an urgent need of the current educational administration system.In the process of teaching quality analysis of educational administration system,the common algorithm is C4.5 decision tree algorithm.However,the accuracy of the results of such algorithms is greatly affected by the selection of initial eigenvalues.At the same time,the size of sample data also has a great impact on the accuracy of the results.This is not conducive to the scientific analysis of teaching quality.In the aspect of algorithm analysis,aiming at the shortcomings of the existing C4.5 decision tree algorithm,this paper introduces convolutional neural network(CNN),which not only realizes the selection and optimization of eigenvalues,but also avoids the accidental error of manual selection by automatically extracting eigenvalues.Firstly,combining with the existing courses,the course attributes(e.g.,the relationship between academic year and specialty,the relationship between semester and class number,etc.)are deeply excavated,and the data of specific periods are extracted as training data samples;secondly,through the introduction of CNN,the initial setting mode of teaching quality decision tree environment is improved.In the original teaching quality analysis algorithm based on decision tree algorithm,by virtue of the advantages of CNN in dealing with non-linear data,the feature extraction process is optimized to obtain more effective initial setting attributes of decision tree analysis features.Through testing,the system can meet the basic requirements of educational administration management in function;meet the basic requirements in performance;in the process of comparative testing of algorithms,this paper introduces CNN optimized C4.5 decision tree algorithm to achieve 94% accuracy on verification set,96% accuracy on test set,and basically achieve the expected.Based on the optimization of teaching quality algorithm and scientific analysis,combined with the actual needs of educational administration management,this paper develops three functional modules: administrator function module,teacher function module and student function module.Through the verification of the system test link,the system achieves the expected effect,and has certain practicability and scientific research value.