电信计费业务中基于复购行为的协同过滤营销推荐

Collaborative Filtering-aided Marketing Recommendation Based on Repurchase Behavior in Telecom Billing Service

移动互联网时代下,通过挖掘电信运营商计费业务订购行为数据,发现用户复购行为规律,并构建用户对于不同业务的兴趣偏好。进一步计算业务之间的相似度,在此基础上提出了基于复购行为的协同过滤营销推荐算法,以提升用户复购频次。为了评估算法的有效性,选取不同的数据模型在视频会员营销中进行测试,结果表明所提算法能够取得较好的转化效果。

In the era of mobile Internet,the rules of users'repurchase behavior,and users'interest preferences for different traffics are investigated by mining the subscription behavior data in telecom billing service.Furthermore,the similarity between applications is calculated,and it proposes a collaborative filtering-aided marketing recommendation algorithm based on the repurchase behavior to improve the frequency of repurchase.In order to evaluate the effectiveness of the algorithm,different data models are selected to test in video member marketing.The results show that the proposed algorithm can achieve better conversion effect.