Analysis Of The Matching Rate And Research On Optimization Techniques Of Chinese College Entrance Preferential Selection And Enrollment

With the rapid growth of information and technology,the emerging expert system has been applied to the guidance of college application filling.The application filling and admission expert system is a personalized recommender system which can target on the students in need and provide punctual service on its own.It is widely used in the cut-off point line prediction,college and specialty recommendation,so there are many college application filling guidance apps in the market.According to research,most of them still exist drawbacks influencing provide scientific recommendations for the students,such as preferences of students ignorance,lack of data,unreasonable college recommendation.There are some problems like specialties are not inconsistent between filling in and enrolling,high score students choose a relatively poor specialties and specialties dislocate in incorrect domains.All of these problems may have negative effects on the development of students,causing educational resources waste.From provincial college entrance examination data by Python,the student application database will be built.On the basis of MBTI personality type and Holland's Self-Directed Search,combining with the specialty characteristics and career planning requirements of colleges and universities to achieve the combination of personality type and specialty and the integration of professional interest and professional category.This paper researches a set of personality type and career self-directed search questionsapplicable to the field of college application,then achieving the optimized recommendation to realize the aim of targeting on the students in need.This paper improves many kinds of algorithms to realize punctual service by applying data mining technology to students' aspiration.In view of the policy innovation of college entrance examination and the surge of affiliated attributes in aspiration filling,the factors affecting the prediction results of the fractional line are collected and sorted out.On this basis,the parameters of the artificial neural network model are continuously optimized.Adapting experience-first strategy to generate decision tree to improve the accuracy of rare samples,the college minimum score prediction accuracy evaluation model will be built.Based on the modified K-means algorithm,an analyzer for college classification will be designed.In view of the different clustering results of different cluster centers selection strategies of K-means algorithm,a minimum spanning tree algorithm is introduced to realize the optimal selection of initial cluster centers and improve the accuracy of the results of the algorithm.Based on the design of professional cell set,the DBSCAN algorithm of grid type is used to generate cluster analysis results of professional type,which reduces the iteration time and improves the efficiency of the algorithm.The design of college application filling and admission expert system realize the suggestion of specialty,career direction and college filling.This system is beneficial to the scientific college application filling and people-oriented application filling mechanism conduction,so that colleges and universities will realize multi-evaluation and multi-selection by rational allocation of higher educational resources.