大数据背景下智能手机APP组合推荐研究

Research in Smartphone APP Combinations Recommendation under the Background of Big Data

面对当下智能手机APP种类繁多、层出不穷的局面,数据提取方案的提出和常见APP组合规律的深度挖掘已成为大数据时代的研究热点。在重新界定不同APP类别间关系度量方式的前提下,推出一套完整的关系型数据提取方案。借助社会网络可视化工具初步发现了不同APP类别间的关系程度及分布,而concor模型为APP组合的多层次划分和推荐提供了可靠的方法学依据。研究发现APP的多层次组合划分实现,对现实生活中经常同时使用的APP类别组的挖掘、划分结果具有较好的解释性和现实意义,为智能手机生产者提供了APP研发方向,并能推进智能生产和生活的发展进程。

Be confronted with existed and emerging varieties of APP,a new data extraction program and deep researches on APP combinations have become a hot topic in big data era.With redefining the relationship measurement method among different kinds of APP,the paper comes up with a complete relational data extraction program.The social network visualization tool helps to find the relationship degree and distribution among different kinds of APP.Concor model provides a trustworthy methodology foundation in multi-level division of APP and its recommendation.The research shows that multi-level division of APP combinations mines the common APP sets in real life and gives good explanatory and practical significance,providing direction for the development of the smartphone APP producers and promoting the development of intelligent life and production process.