A novel algorithm for visualizing concept associations

An associative concept space is a map that visualizes the associations between concepts in a scientific field. In this paper, we propose a novel algorithm for constructing an associative concept space. This algorithm can be seen as an alternative to multidimensional scaling, which is typically used in the literature on knowledge domain visualization. We describe experiments in which the proposed algorithm and multidimensional scaling are both used for constructing an associative concept space of the computational intelligence field. It turns out that the associations between concepts in this field are better reflected in the concept space constructed using the proposed algorithm than in the concept space constructed using multidimensional scaling.