Clustering on Social Web
Tomás Kuzár · 2013
Social web increases its potential rapidly. Growing numberofinvolvedusersleadstosignificantincreaseinamount of user-generated content. End users have great opportunity to express themselves by publishing statuses, blogs or photos and in the meantime they consume the content generated by others. In our work we focus on process of social web data consumption- gathering, processing and visualization. In our research we focus on processing of unstructured textual content of Social Web in order to achieve more efficient access to relevant information. We have designed and evaluated methods for building precise content clusters by mining social web data. Our findings indicate the need to encounter external knowledge and the internal relationships between objects on social web to increase the accuracy of extracted knowledge. In user study we demonstrate how the accurate content clusters augment the access to relevant information on the social web.