In practice, analysts need to apply their domain knowledge to examine the algorithm output. To address the issues of the related art, a generic visual analysis framework for detecting and examining missing links in bipartite networks is proposed in the present disclosure. First, the framework contributes a novel link prediction approach for bipartite networks, which is an ensemble method leveraging the information of bicliques in the networks. Second, an interactive visualization is utilized to present detected missing links and allow for a better understanding of the meaning and influence of missing links, through two of the most common network analysis approaches: metric-based (e.g., computing node betweenness) and motif-based (e.g., detecting cliques).

Further, no related art system addressed the problems of detecting and visualizing missing links. More particularly, in example implementations, a matrix-based design is employed because links are the focus in our framework and need to be emphasized visually.