The role of social tags in web resource discovery: an evaluation of user-generated keywords

Abstract

Social tags are user generated metadata and play vital role in Information Retrieval (IR) of web resources. This study is an attempt to determine the similarities between social tags extracted from LibraryThing and Library of Congress Subject Headings (LCSH) for the titles chosen for study by adopting Cosine similarity method. The result shows that social tags and controlled vocabularies are not quite similar due to the free nature of social tags mostly assigned by users whereas controlled vocabularies are attributed by subject experts. In the context of information retrieval and text mining, the Cosine similarity is most commonly adopted method to evaluate the similarity of vectors as it provides an important measurement in terms of degree to know how similar two documents are likely to be in relation to their subject matter. The LibraryThing tags and LCSH are represented in vectors to measure Cosine similarity between them.