Higher Vocational Students' Employment Based On Decision Tree Classification Algorithm Analysis And Prediction

This thesis was based on the background of the requirement for the quality of higher vocational education in "the National Long-term Education Reform and Development Plan(2010-2012)" and the classification analysis algorithm of data mining and machine learning.The some years of enrollment data, students enrollment management data and employment data of Yunnan Vocational College of Mechanical and Electrical Technology were regarded as the samples in this thesis. The prediction model for predicting whether the graduates can successfully obtain employment and the types of the employment units was proposed. The prediction model works by analyzing the information of the students admissions test scores, the evaluations of comprehension quality and the region and the properties of the employment units, and the information is obtained by learning the decision tree classifier and analyzing the factors of employment.This thesis was based on the some years of enrollment data, students enrollment management data and employment data of Yunnan Vocational College of Mechanical and Electrical Technology, combined the current situation and the characteristic of the employment among the students of Yunnan Province vocational colleges, preprocessed the relevant data that influence the students employment, analyzed the employment decisions. Using the ID3classification algorithm, the decision tree model of predicting employment, the rule of prediction and the exact evaluations were generated. Then the employment data of the graduates were grubbed and analyzed, the regular potential information was obtained, the prediction model for predicting whether the graduates can successfully obtain employment and the types of the employment units was generated..Classifying and predicting the employment of the graduates according to the students admissions test scores, the evaluations of comprehension quality and the region and the properties of the employment units can obtain the rule of prediction and lay the foundation for further analysis and applications.The experimental results show that the prediction model which is proposed by the decision tree classifier ID3algorithm is accurate and efficient.Overall, the main work of this thesis can be summarized as the following:(1) Based on the present study, Transforming the enrollment data, students enrollment management data and employment data into the decision attribute of the employment prediction.(2) Establishing the complete model for predicting whether the graduates can successfully obtain employment and the types of the employment units by choosing the decision attributes.(3) The analysis of the model, the experimental testing and the performance analysis results showed that the prediction model which was proposed by this thesis had a high accuracy.