孙克雷,邓仙荣.一种改进的基于梯度提升回归算法的O2O电子商务推荐模型[J].安徽建筑大学学报,2016,24(): |
一种改进的基于梯度提升回归算法的O2O电子商务推荐模型 |
O2O E-commerce Recommendation Model on improved Gradient Boosting Regression Tree |
投稿时间:2016-01-07 修订日期:2016-04-18 |
DOI: |
中文关键词: 梯度提升回归树,位置服务,个性化推荐,行为日志分析 |
英文关键词: GBDT, LBS, Personalized recommendations, behavior log analysis |
基金项目:安徽省自然科学基金 |
|
摘要点击次数: 3488 |
全文下载次数: 1216 |
中文摘要: |
位置属性对于线下消费的用户具有重要影响。为了有效提高个性化推荐精度,在对O2O电子商务特点进行用户特征分析的基础上,在推荐算法中引入当前时间参数和位置参数,提出了一种改进的基于梯度提升回归算法的O2O电子商务推荐模型。实验结果表明,改进的基于梯度提升回归算法的O2O电子商务推荐模型在实时性和准确性方面明显优于传统的推荐算法。 |
英文摘要: |
Location attribute has important influence for offline consumption of users. In order to improve the accuracy of personalized recommendation, on the analysis of the O2O e-commerce with user characteristics, it introduces the current time and location parameters to the basis of the recommendation algorithm , and it is proposed based on an improved gradient boost O2O e-commerce recommendation model of regression algorithm. The experimental results show that the improved improve regression algorithm based on gradient O2O e-commerce recommendation model in real-time and accuracy is superior to the traditional recommendation algorithm. |
查看全文
查看/发表评论 下载PDF阅读器 |
关闭 |