Research On The Application Of Data Mining Technology In Analysis Of Examinee Wish

With the rapid development of database technology and widely usage of database management system, a great deal of historical data in various trades and industries have been accumulated, and that the increasing historical data often conceal a lot of important information. It is an important research topic how to timely detect useful knowledge from historical data and to excavate the value of its potential to increase their utilization. As part of its solution, in recent years, data mining (Data Mining, DM) technology is rising rapidly.To the province as a unit, millions of correlative will records of annual college entrance examination information is stored distributedly in multiple databases of every areas. These large-scale data contain the rich decision-making information and knowledge. Developing these valuable information resources to serve the college entrance examination enrollment and to guide candidates to fill wills availably is an important task,and is one of the issues need urgently to solve.This paper takes the data mining technology as the data analysis method, according to the college entrance examination wish data's characteristic and self-localization principle of the parallel wish throw file, carries on the analysis and the research to the data of college entrance examination wish and its correlative data, proposes a risk assessment model of examinee wish based on the data mining technology, and gives the corresponding concrete algorithm and solution.The contents are as followed:1. The data of college entrance examination wish and its correlative data be carried on the pretreatment.2.The paper uses flexibly the decision tree C4.5 algorithm and makes the improvement, to contrapose the shortcoming to neglect easily attributes of the little data and to increase the scarce sample precision, makes use of associated method of the optimization of the attribute recursion and the optimal strategy based on experience;at the same time, proposes the C4.5 algorithm's processing mode who can carry on the increase study, produces the decision tree under the non-balanced data set, designs a classifier of the wish analysis.3.According to the actual situation of the self-localization when candidates fill wills,the paper chooses reasonably the initial centers of K-Means clustering algorithm,uses the K-Means clustering algorithm to carry on the clustering with the colleges and universities,carries on the intellective excavation to the correlative college entrance examination data, gains regular latent information.4.The paper establishes a risk assessment model of examinee wish,carries on the forecast to situations of the college entrance examination enrollment according to examinee's achievement, sort of the subject, register specialty,situations of the colleges and universities and so on, obtains that each wish have many colleges and universities which can be enrolled and risk coefficients of this colleges and universities wish in the parallel wish, provides examinees who fill the wish during the college entrance examination.Based on the above theory,the developed system called "Risk Assessment System of Examinee Wish" has experimentized on data of the college entrance examination's wish information of Henan Province of last three years.The result is inosculated with the distribution of the college entrance examination actual matriculate information of that year.Therefore,it is really sure that the testing results,as the effective reference value,are good to examinee of the coming year for filling in wish.