Social Semantics and Similarities from User-generated Keywords to Information Retrieval: A Case Study of Social Tags in Marine Science

Abstract

Of late, social tagging has become popular trend in information organisation. In context of digital resources the tags assigned by users also play vital role in information retrieval. For information discovery the ‘terms’ used to retrieve the results also depend upon the ‘relevancy’ or ‘weightage’ of the keywords. This study investigates ‘relevancy ranking’ of terms used in the full text of the resource. The common words present in both full text of the article and social tags were considered for the study by employing TF-IDF statistical technique and Jaccard similarity test. The results show that it is possible to assign ‘weight’ to keywords for better results and also determine the significant tags assigned by the user. The Jaccard similarity coefficient test adopted to understand the word similarity between full text words of an article and marine social tags. This work reveals the social tags can enrich metadata for information retrieval.