Towards a Correlation Cooccurrence Model Generating Approach to Folksonomy

Abstract

Folksonomy is a hyper graph data structure in social network, in which co occurrence is often used to act as an important means of recommender system. The co occurrence data information from folksonomy hyper graph is becoming increasingly important in recommending applications of social network. This paper presents an original method for easy random approach to generating correlation co occurrence model of folksonomy hyper graph in social network from a random user. The method creates classified co occurrence patterns, using resource items, tags, or users, correlation co occurrences obtained from two given corpora. Our experiments demonstrate that proposed approach can produces better stability for the recommender system. And our method offers a feasible means for developers to handle information co occurrence problems for folksonomy application.