Link prediction in linked data of interspecies interactions using hybrid recommendation approach

Abstract

Linked Open Data for ACademia (LODAC) together with National Museum of Nature and Science have started collecting linked data of interspecies interaction and making link prediction for future observations. The initial data is very sparse and disconnected, making it very difficult to predict potential missing links using only one prediction model alone. In this paper, we introduce Link Prediction in Interspecies Interaction network (LPII) to solve this problem using hybrid recommendation approach. Our prediction model is a combination of three scoring functions, and takes into account collaborative filtering, community structure, and biological classification. We have found our approach, LPII, to be more accurate than other combinations of scoring functions. Using significance testing, we confirm that these three scoring functions are significant for LPII and they play different roles depending on the conditions of linked data. This shows that LPII can be applied to deal with other realworld situations of link prediction.