Semantic relatedness (SR) is a form of measurement that quantitatively identifies the relationship between two words or concepts based on the similarity or closeness of their meaning. In the recent years, there have been noteworthy efforts to compute semantic relatedness between pairs of words or concepts by exploiting various knowledge resources such as linguistically-structured（e.g., WordNet) and collaboratively-developed knowledge bases (e.g., Wikipedia), among others. The existing approaches rely on different methods for utilizing these knowledge resources, for instance, methods that depend on the path between two words, or a vector representation of the word descriptions. The purpose of this paper is to review and present the state of the art in semantic relatedness research through a hierarchical framework. The dimensions of the proposed framework cover three main aspects of semantic relatedness approaches including the resources they rely on, the computational methods applied on the resources for developing a relatedness metric, and the evaluation models that are used for measuring their effectiveness. We have selected fourteen representative SR approaches to be analyzed using our framework. We compare and critically review each of them through the dimensions of our framework , thus, identifying strengths and weaknesses of each approach. In addition, we provide guidelines for researchers and practitioners on how to select the most relevant SR method for their purpose. Finally, based on the comparative analysis of the reviewed relatedness measures, we identify existing challenges and potentially valuable future research directions in this domain.

Keywords