基于运营商大数据的APP潜在用户挖掘研究

Research on APP Potential User Mining Based on Big Data of Operator

跨界合作是运营商增加后向收入和提高用户黏性的重要途径,而实现降本增效的关键在于对目标用户的精准圈选。着重关注互联网APP的合作推广,提出基于上网日志数据构建潜在用户模型的方法,并对4种机器学习算法进行比较,发现LightGBM模型效果最好,且远优于业务规则模型。

Cross industry cooperation is an important way for operators to increase backward revenue and improve user stickiness,and the key to reducing cost and increasing efficiency is to select target users accurately.Focusing on the cooperative promotion of the Internet APP,the method of constructing potential user model based on Internet log data is proposed,and four machine learning algorithms are compared.It is found that the LightGBM model has the best effect and is much better than business rule model.