Application Of Folksonomy In Government Information Portal

Government website is an important part of E-government,and the government also provides the window face of management and services to the public. Personalized and service government website has been expected by everyone. The information organization and classification system is a hot issue in government management. Folksonomy is a kind of new resources organization method which appears in the age of Web2.0, which helps generate community of users and resources, cut off the cost of information retrieval, and has semantic and top-hot characters.This paper introduces the Folksonomy and related concepts of government websites, and by analyzing the existing information classification model of the government websites at home and abroad to study the traditional classification problems. Based on these contents, this paper proposed a government based on Folksonomy portal subject-public classification system, the establishment of a personalized website for the government classification system. On this basis, against government site content without the level of features made popular tracking system, this way users can be hot topics of interest to outline the different levels of the surface in the government portal website. Finally, for the tag clustering problem of subject- public classification system, the introduction of the well-known K-means clustering method, and the algorithm was proposed to improve the K-means algorithm is based on the label clustering algorithm, which removes tags tail noise, increase the semantic category tags.