Application Of Fuzzy Clustering In Undergraduate Engineering Quality Certification

Undergraduate engineering quality certification has important guiding significance for the teaching reform and teaching management of colleges and universities.Choosing an appropriate evaluation method plays multiplier role in undergraduate engineering quality certification in universities.The undergraduate engineering quality certification work is carried out from two aspects: teaching level and student learning achievements.T Therefore,this paper analyzes the data from the two aspects of teacher teaching level and student learning achievements.When evaluating the quality level of colleges and universities,it is usually based on class rate and student performance,but these evaluation methods are inevitably one-sided and cannot accurately reflect the teacher's comprehensive teaching level and students' various abilities.Therefore,this article combines quantitative and qualitative methods to study and analyze the application of fuzzy clustering in undergraduate engineering quality certification from various aspects.Fuzzy clustering method(FCM)is an important analysis method in data mining technology.Because of its simple principle,easy implementation and fast convergence,it is widely used in various industries.The number of clusters of the FCM algorithm needs to be set in advance and the initial center is also randomly generated.Generally,the clustering result will prematurely converge,and the final clustering results are not ideal.Therefore,this paper combines the FCM clustering algorithm with the genetic algorithm.To achieve the final convergence result of the FCM clustering algorithm to reach the global optimum,finally the GA-FCM algorithm was applied to the undergraduate engineering quality certification.This article introduces the relevant theories of data mining technology.It focuses on the principles and implementation processes of the classic clustering algorithm and the FCM clustering algorithm and analyzes the advantages and disadvantages of classical clustering algorithm and fuzzy clustering algorithm.In order to make the FCM clustering algorithm result better,choose to use genetic algorithm to improve its defects,and use the standard data set in the UCI machine learning library for experiment and result analysis,which proves that it can effectively improve the clustering result.Secondly,a questionnaire survey was used to evaluate and analyze the teacher's level.Data were collected from five aspects: basic ability,teaching method,teaching content,teaching attitude,and teaching effect.The third-party library pandas in python3 was used for data cleaning.The applied data was applied to the fuzzy K-means clustering algorithm based on genetic algorithm,and the results were analyzed,so as to provide relevant data for improving the teaching work for teachers of electrical majors in the School of Electrical and Computer,Jilin Jianzhu University.Finally,relying on the "Jilin Architecture University Engineering Professional Platform" analysis and evaluation of students' learning results.The data of each subject's scores are processed according to the weight of each course in the graduation requirements,and the data are prepared using the 12 graduation standards of electrical undergraduates.Finally,clustering and result analysis are performed,which can guide teachers in For students with different characteristics in teaching,choose appropriate teaching methods,maximize the advantages of students,make up for shortcomings,and finally enable all students to meet the graduation requirements and become strong comprehensive engineering and technical personnel to meet social needs.