Network-based link prediction of scientific concepts – a Science4Cast competition entry

  title={Network-based link prediction of scientific concepts – a Science4Cast competition entry},
  author={Jo{\~a}o P. Moutinho and Bruno Coelho Coutinho and Lorenzo Buffoni},
  journal={2021 IEEE International Conference on Big Data (Big Data)},
We report on a model built to predict links in a complex network of scientific concepts, in the context of the Science4Cast 2021 competition. We show that the network heavily favours linking nodes of high degree, indicating that new scientific connections are primarily made between popular concepts, which constitutes the main feature of our model. Besides this notion of popularity, we use a measure of similarity between nodes quantified by a normalized count of their common neighbours to… 


Predicting missing links via local information
A new similarity measure, motivated by the resource allocation process taking place on networks, is proposed and shown to have higher prediction accuracy than common neighbours, which can remarkably enhance the prediction accuracy.
The link-prediction problem for social networks
Experiments on large coauthorship networks suggest that information about future interactions can be extracted from network topology alone, and that fairly subtle measures for detecting node proximity can outperform more direct measures.
Link Prediction in Complex Networks: A Survey
Quantum Link Prediction in Complex Networks
A quantum algorithm for link prediction, QLP, which uses quantum walks to infer unknown links based on even and odd length paths, and is at least as precise as state-of-the-art classical link prediction methods, both in cross-validation tests and in the prediction of experimentally verified protein-protein interactions.
Link prediction in social networks: the state-of-the-art
A systematical category for link prediction techniques and problems is presented, and some future challenges of the link prediction in social networks are discussed.
Degree distributions of bipartite networks and their projections
This work uses the formalism of generating functions to prove that when projecting a bipartite network onto a particular set of nodes, the degree distribution for the resulting one-mode network follows the distribution of the nodes being projected on to, but only so long as the degree distributions for the opposite set of node does not have a heavier tail.
Networks: An Introduction
This book brings together for the first time the most important breakthroughs in each of these fields and presents them in a coherent fashion, highlighting the strong interconnections between work in different areas.
What is network science?
The first few pages of the first issue of Network Science, the new journal created because network science is exploding, share their own vision of the emerging science of networks.
Friends and neighbors on the Web
Network-based prediction of protein interactions
It is shown that proteins tend to interact if one is similar to the other’s partners and that PPI prediction based on this principle is highly accurate and can offer mechanistic insights into disease mechanisms and complement future experimental efforts to complete the human interactome.