Persona Based WordPress Blog Posts Recommendation Research

Open source software is favored by more and more software developers because of the advantages of low cost,easy to use and easy maintenance.WordPress,an open source software to create blogs,has become the most widely used blogging software in the world since launched.But the related studies are all about how it can be used to build blogs,the vast amount of historical data left on the platform by users is wasted.It is noteworthy that:now blogs built by WordPress can not realize personalized recommendations,seriously affecting the user experience.In order to solve the above problems,this paper takes WordPress as the research object and uses literature analysis,description statistics and case studies to analyze and summarize the literature from three aspects:persona construction,precise recommendation algorithm and diversity recommendation algorithm.First,the general process of constructing user portrait is proposed,user portrait labels and data collection indicators are designed,and the data source is determined.At the same time,we added user behavior time weight and user behavior dynamic weight when using project-based K-nearest neighbor algorithm to model user's precise preference.User's basic attributes were used to calculate user's similarity,and users' potential preferences were modeled according to similar users.Then,it designed the user portrait database and the data processing scheme for WordPress,and developed the blog recommendation and plug-in recommendation plug-in to achieve WordPress blog recommendation.Finally,the historical data of users in the English blogs focusing on Chinese culture are extracted,cleaned up and transformed,and plugins are installed and configured in the blog to verify the feasibility and the effectiveness of the proposed program.On one hand,this thesis enriches the theories,methods and techniques of persona based WordPress blog posts recommendation research.On the other hand,it improves the recommendation function of WordPress blog to meet the personalized reading requirements of users by developing plugins.