论文部分内容阅读
大数据时代,互联网行业内得数据者得天下,搜狐也自2014年成立大数据中心,在PC端,搜狐已覆盖中国90%的网民,并以多元化平台搜集多元化数据。但仅仅拥有丰富且多元的数据并非搜狐的唯一优势,依托于数据转化、数据算法形成的用户标签也是搜狐的另一优势。基于大数据算法,描绘用户的物理属性、内容行为、广告行为构成十大兴趣标签及多级标签体系。
In the era of big data, those in the Internet industry have access to the world’s data. Sohu has also set up a big data center since 2014. On the PC side, Sohu has covered 90% of Internet users in China and has diversified platforms to collect diversified data. However, having rich and diverse data is not the only advantage of Sohu. Relying on data conversion, the data tagging algorithm is another advantage of Sohu. Based on the big data algorithm, depicting the user’s physical attributes, content behavior, advertising behavior constitutes the top ten interest labels and multi-level label system.