Research On The Application Of Folksonomy In Information Resources Organization Of Medical Portal Website

The rapid development of the Internet and the huge demand for medical information resources have promoted the rise of medical portals.Medical portal has become an indispensable platform for people to understand the information of medical and health resources.With the increasing demand for medical health information,the organization and classification of information resources of medical portal websites are becoming the focus of attention.As an important product of the Web2.0 era,Folksonomy has great advantages in traditional information classification,which greatly reduces the classification cost of information resources,effectively shortens the updating cycle of categories,which strong semantic and tagged features will greatly influence the improvement of the organizational structure of the information resources of the medical portal.This thesis first introduces the related concepts and characteristics of the Folksonomy,analyzes the main models of the information classification of medical portal websites,points out some shortcomings and challenges in the current classification model,while Folksonomy in the classification of information resources,which advantages will help to solve these problems,so this paper put forward a medical portal classification system based Folksonomy,the new model make the original classification system better,so that the information organization system towards personalized,dynamic direction,The maximum to meet the needs of users.In this paper,K-means clustering algorithm is used to resolve the problem of how to combine Folksonomy with the traditional classification model.In order to achieve better clustering effect in the medical portal,this paper improves the original algorithm,clustering the topic words and tags according to the similarity,and generates semantic strong subject-tag set tree structure,Through simulation experiments,the paper use authoritative evaluation index system to verify the advantages and disadvantages of clustering on the subject-tag tree set.