Design And Implementation Of Blog Recommendation System Based On Parallel Frequent Pattern Mining Algorithm

With the rapid development of Internet technology,our lives are full of various types of data.In recent years,as a product of social platforms,blogs have attracted more and more people to write blogs online to record their learning status.The current homepage of a blog site is usually a popular blog on the Internet.If you want to recommend a certain field,you need to pay attention to a column or blogger in the field,and then recommend their blog.Rarely consider starting from the collection function of the blog to research personalized recommendations.In this paper,the frequent pattern mining algorithm is combined with MapReduce to mine the recorded data of the blog.According to the above-mentioned blog site's defieiencies,a personalized,recommendation is proposed from the collection function of the blog.It is to find blogs associated with the user's favorite blogs in the frequent items of the blog's record resources and sort them by matching processing,and then recommend them to the user.In order to visually display the personalized recommendation function of the blog,this article uses the SSH framework to design and implement a blog recommendation system.This article adds algorithm running and task monitoring functions to the basic functions of the blog,and can remotely invoke the Hadoop cloud platform online operation algorithm.At the same time,the data after the algorithm is run is parsed and downloaded to the local database.According to this data file and the user's favorite blog,personalized recommendations are made.In this paper,under the background that many scholars use collaborative filtering as the recommendation algorithm,the frequent pattern mining algorithm is used to study the recommendation system,a personalized recommendation algorithm model is proposed,and the recommendation results are displayed with the blog system.