随着市场竞争日益激烈和信息化技术不断发展,通过数据分析和挖掘来预测新的潜在商机成为了企业商机管理的重要环节.现有机器学习算法主要基于样本数目趋于无限大的假设,但实际问题中样本大多是有限的,甚至是小样本数据,难以保证机器学习结果的合理性.将支持向量回归(SVR)算法用于商机预测建模过程,用于解决小样本、高维数、非线性的学习问题.实验结果表明,与决策树等算法构造的目标函数求解结果相比较,SVR算法在有限样本空间能获得较高精度的预测结果.%With the increasingly fierce market competition and the development of information technology,the prediction of potential business opportunities through data analysis and data mining becomes an important part of business opportunities management.Most of machine learning algorithms are principally based on the hypothesis that the number of samples tends to be infinite,but the reality is different so that the reasonableness of the results cannot be guaranteed.Support vector regression (SVR) algorithm is used to predict the reliability of the data modeling process to address small sample,multi-dimension,nonlinear problems of the training model.The results show that SVR algorithm has a high accuracy of predicting results in limited sample space.
展开▼
