| 融合用户画像与协同过滤的知识付费平台个性化推荐模型 |
| |
| 引用本文： | 魏玲,郭新悦.融合用户画像与协同过滤的知识付费平台个性化推荐模型[J].情报理论与实践,2021(3):188-193. |
| |
| 作者姓名： | 魏玲 郭新悦 |
| |
| 作者单位： | 哈尔滨理工大学经济与管理学院 |
| |
| 摘 要： | 目的/意义]为提高知识付费平台用户感知服务质量,文章构建了融合用户画像与协同过滤的个性化推荐模型。方法/过程]首先根据用户特性构建画像标签体系,利用TF-IDF、熵值法、k-means等方法确定用户特征标签;其次分别基于用户画像与改进后的协同过滤算法计算用户相似度,通过调和权重得到用户综合相似度;最后利用Top-N进行个性化推荐。结果/讨论]通过知乎live付费用户信息进行验证,发现本文算法在推荐结果的准确率以及召回率上,相比其单一方法均有较大提升,且满意度高于知乎live平台。
|
| 关 键 词： | 用户画像 知识付费平台 个性化推荐 协同过滤 |
Personalized Recommendation Model of Knowledge Payment Platform Combining User Portrait and Collaborative Filtering |
| |
| Abstract: | Purpose/significance]This paper builds a personalized recommendation model that combines user portraits and collaborative filtering,aimed at improving the perceived service quality of knowledge-paying platform users.Method/process]First,build a portrait label system based on user characteristics,use TF-IDF,entropy method,k-means and other methods to determine user feature labels;secondly,calculate user similarity based on user portraits and the improved collaborative filtering algorithm.The comprehensive similarity of users is obtained by reconciling weights;finally,Top-N is used for personalized recommendation.Result/conclusion]Through the verification of Zhihu live paid user information,it is found that the precision and recall rate of the recommended results of the algorithm in this paper are greatly improved compared to its single method,and the satisfaction is higher than that of the Zhihu live platform. |
| |
| Keywords: | user portrait knowledge payment platform personalized recommendation collaborative filtering |
| 本文献已被 维普 等数据库收录！ |
|
