【摘要】 将自适应提升方法(AdaBoost)应用于二手车价值的评估,提出一种以决策树桩作为弱分类器的集成方法。二手车数据样本量大,实体特征多,通过区间离散化得到样本集,避免深度遍历产生过拟合。通过加权表决集成弱分类器,建立分类模型。实验表明,AdaBoost方法相比传统的决策树方法,准确率提高7.1%。更多还原

【Abstract】 To apply adaptive boosting(AdaBoost) to the evaluation of used car value,a boosting method of decision stump as a weak classifier is proposed. Because of the large quantities of sample data and features,the sample set is got through interval discretization to avoid the over-fitting due to the depth of traversal. The weak classifier is integrated by weighted voting to establish the classification model.The experiments prove that the accuracy of the Ada Boost is 7. 1% higher than that of the traditional decision tree.更多还原