Analysis Of College Employment Data Based On Decision Tree Algorithm

According to the data released by the National Bureau of Statistics,the gross enrollment rate of higher education in China reached 40% in 2015,according to the theory of the division of higher education development stage proposed by Professor Martin Trow.China's higher education has entered the stage of mass education.Although the college's13-year undergraduate enrollment has stopped,the number of college graduates is increasing every year.In 2017,the number of college graduates reached 7.95 million,and the number of graduates in 2018 has exceeded 8 million.The number of college graduates has exceeded 7million for three consecutive years.Since the implementation of reform and opening up,the number of colleges and universities in China has increased.Under the new era of employment development,China's various industries are developing and attaching importance to talents.In particular,talent has increasingly become the top priority of national development.The number of graduates is increasing year by year,the job market is becoming more professional,liberalized,and market-oriented,and the difficulty of employment seems to be the norm.Colleges and universities have also been helping graduates to increase employment rates,such as holding multiple job fairs and multiple job guidance lectures,but the results are average.Nowadays,the society is in the era of big data.Data mining in big data clock is a widely used technology today.Data developers use data mining technology to explore data deeper and explore the relevance of data and discover the correlation between data to provide decision support for decision makers.After the realization of college informationization,With the continuous maturity of data mining technology,it has been widely used in the financial field,telecommunications industry,retail industry,medical telecommunications field,transportation field,etc.,and the application of college education in this aspect is still lagging behind,especially in colleges and universities.Less is used in it.In the employment information data of college students,there are many years of student employment information,and the employment information data of these graduates is closely stored in the computer hard disk,and at most,it is used by the employment management personnel for ordinary information inquiry,and the utilization rate is very low.Give full play to the potential power of employment data of colleges and universities,make employment guidance work more efficient,more accurate,more individualized,and more targeted to serve students.At the same time,it provides reference for college employment services and promotes new measures for college employment.This paper compares and analyzes several typical decision tree algorithms,focusing on the ID3 algorithm and C4.5 algorithm.The previous decision tree algorithm of the ID3 algorithm requires that all values of the data set be accurate.Missing data will reduce the performance of the algorithm.If the missing data is not handled properly,a large number of errors will be accumulated,which will increase the computation time and complexity of the subsequent algorithms.Although the C4.5 algorithm can handle missing data,it is still not perfect.The decision tree algorithm is optimized using association rules and Bayesian algorithms.The missing values of the decision tree are filled with the Yes model,and then the decision tree algorithm is used to construct the decision.The data was obtained from the 2016 graduate data of a college in Jingdezhen.This article is mainly from five parts.The first part introduces Introduce research background,content,framework,meaning and method,as well as the current status of data mining technology in college employment.The second part introduces the basic concepts of data mining,as well as the concept of decision tree algorithm and the typical algorithm of decision tree.In the third part and the fourth part,the decision tree algorithm is optimized by association rules and Bayesian rules,and the data is used for testing,verification and analysis.The fifth part is based on the decision tree algorithm of employment data analysis system design,analyzes the student employment information data of a college in Jingdezhen,and verifies the reliability of the algorithm.Finally,the summary and outlook of the full text.Mining employment data related to college students,discovering potential laws,finding hidden models,and providing decision-making basis for employment guidance,thus promoting the reform of graduate employment system and promoting employment of college students.