Similarity measures are essential to solve many pattern recognition problems such as classification, clustering, and information retrieval. Various similarity measures are categorised in both syntactic and semantic relationships. In this paper, we present a novel similarity, unilateral Jaccard similarity coefficient (uJaccard), which does not only take into consideration the space among two points but also the semantics among them. How can we retrieve meaningful information from a large and sparse graph? Traditional approaches focus on generic clustering techniques for network graph. However, they tend to omit interesting patterns such as the paradigmatic relations. In this paper, we propose a novel graph clustering technique modelling the relations of a node using the paradigmatic analysis. Our proposed algorithm paradigmatic clustering (PaC) for graph clustering uses paradigmatic analysis supported by an asymmetric similarity using uJaccard. Extensive experiments and empirical analysis are used to evaluate our algorithm on synthetic and real data.