Research On The Accurate Help Of The Students In Poverty Based On Apriori-DT

As a factor restricting economic and social development,the problem of poverty is highly valued by all countries in the world.From "accurate poverty alleviation" is proposed in our country,to the poverty alleviation and development "in precise,accurate,success or failure of is accurate" implementation of "precise poverty alleviation" has become an important strategic task in our country.Poverty alleviation is both "FuZhi" and "FuZhi",colleges and universities as an important base of personnel training in our socialist construction,how to overcome the "flood irrigation" way of poverty alleviation and realize the accurate support has become an important task to university administrators.Based on this,the research is a combination of data mining methods,explores the main factors of accurate evaluation of the poor students,build the accurate classification of the decision tree model,in order to the sponsors to implement precise support work put forward feasible countermeasures and Suggestions.This study first sorted out the theories and researches on the identification and data mining of poor students at home and abroad.Secondly,based on the data of the poor students and the data of guizhou students,the data of the poor students in guizhou colleges and universities were compared and screened.At the same time,the Apriori correlation algorithm and C4.5 decision tree algorithm were used to extract the main features of the poor students,and the characteristics were regrouped into new attributes to achieve the reduction of the original index.Finally,70% of the data is randomly selected as the training set to construct the accurate classification decision tree model for poor students in colleges and universities,and the remaining 30% data is used as the test set to verify the model.Research were extracted by family per capita net income of less than 2300 yuan,students to apply for student loans,family population more than 10 people,head of the household culture degree is elementary school and junior high school,students are not out of poverty and orchard area composed of 12 indexes such as less than 0.39 acres of seven major characteristics.This paper constructs a classification decision tree for guizhou poor students whose income per capita is less than 2300 yuan and students apply for student loans as the first branch,and the classification effect is verified.The results show that the overall forecast accuracy is 84.5%,and the classification effect is better.Based on the above-mentioned classification decision tree for poor students,this paper provides help countermeasures and suggestions for students with different poverty levels from the two aspects of " helping the intellect" and " helping the mind".In order to promote Guizhou's university aid workers to achieve accurate assistance to poor students.