Research On Accurate Push Of Tik Tok Information Based On User Portrai

In the context of big data,the growing popularity of short videos has brought about tremendous changes in people’s daily lives,as well as numerous development opportunities for various industries.In the process of short video content delivery,accurate delivery services meet the diverse information needs of users.By adopting customized and intelligent information delivery systems,high-speed and precise matching between users and data is achieved,greatly reducing the difficulty of data transmission and information collection,which brings convenience to users’ work and life.User portraits have become an important component in the process of precise short video information delivery.Based on this,this article selects some Tiktok post-00 s as the research objects to construct and analyze their user portraits,mine the potential interests and preferences of the group,in response to the existing issues in the precise information push process of Tiktok,we propose the optimization of Tiktok’s information precise push strategy based on the Golden Circle thinking.Firstly,this article provides an overview of the current research status,significance,and methods of user profiling and targeted information delivery in both domestic and international contexts.Secondly,this article outlines relevant theories such as user portraits and user life cycle.Based on this,this article selects Tiktok as the research target,uses Python crawler technology to collect and process data information of post-00 s user groups on Tiktok,such as basic characteristics,behavioral preferences,and interest preferences,and labels the data information to form a user portrait that defines and quantifies post-00 s user groups,aiming to mine potential preferences of target users.The research shows that most of the sampled post-00 s user groups are concentrated in first-tier or new first-tier cities such as Beijing,Shanghai,Guangzhou,Shenzhen,and Chengdu;male users occupy the majority of the user group;the age distribution is mostly concentrated between 16 and 22 years old;educational distribution mainly focuses on undergraduate and college degrees;the active time period is usually from 10 am to 2 pm and from 4 pm to 10 pm;users who love science experiment videos have the most fans.Furthermore,this article conducts research through Tiktok’s related user portraits and their reflected needs,and simulates the accurate delivery results divided by type ratios.The results show that the method based on user portraits can accurately extract relevant characteristics of target user groups,which helps to obtain more accurate recommendation results in the process of accurate information delivery on Tiktok.Finally,this article points out the problems in the content precise delivery of short video platforms,such as low-quality content delivery and big data discrimination,and proposes corresponding improvement paths and optimization strategies for the problems.