Open AccessPrediction of Poor Students’ Classification Based on Adaboost Algorithm Integrated Learning Model
Author(s) -
Wang Li,
Guie Jiao
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1574/1/012172
Subject(s) - adaboost , undersampling , machine learning , computer science , artificial intelligence , decision tree , classifier (uml) , ensemble learning , decision tree learning , principal component analysis , algorithm , pattern recognition (psychology)
Aiming at the problem of identifying the poor students in colleges, based on the improved classification imbalance, this paper proposes to establish the AdaBoost algorithm integrated learning model for the first time to solve the problem of identifying the poor students in colleges. In this model, decision tree is used as weak learner, and principal component analysis algorithm and balance cascade algorithm are used to reduce dimension and undersampling data. At the same time, the paper focuses on the comparison of the algorithms that choose the effect of sample set classification as the weight(SAMME.R) and the size of the prediction probability of sample set classification as the weight(SAMME). Experiments have proved that the AdaBoost ensemble learning algorithm model is better than the single classifier algorithm in predicting the results, which can ensure the fairness and objectivity of the recognition results, and has a strong application value for the recognition of poor students in colleges.