Knowledge Level Tags: Applied to Collaborative Recommender Systems on the Web

Tracking #: 3512-4726

This paper is currently under review
Bruno Zolotareff dos Santos
Jorge Rady de Almeida Junior
Sandra Santos Vales

Responsible editor: 
Axel Polleres

Submission type: 
Full Paper
This article aims to present a tag recommendation model at the knowledge level in a collaborative system on the Web. One of the main reasons for this proposal is due to limitations in the tagging process, causing loss in the quality of the terms used in the metadata that are indexed in posts on social networks in the form of tags, losing the meaning of the relationship between the tags and the object, resulting in a lack of engagement in the collaborative system by not exploring the potential of collective intelligence in a more practical and visual way to be identified by the user when choosing tags in the process of indexing the object. In this study, an algorithm for classifying metadata at the knowledge level is proposed, which uses metrics capable of measuring the collective intelligence aggregated to the metadata generated in the system, with two main steps being assigned, which are the classification and recommendation of a set of tags at the knowledge level.
Full PDF Version: 
Under Review