https://doi.org/10.1140/epjs/s11734-024-01145-y
Regular Article
Predicting hyperlinks via weighted hypernetwork loop structure
1
Key Laboratory of Intelligent Education Technology and Application of Zhejiang Province, Zhejiang Normal University, Yingbin Avenue, 321004, Jinhua, Zhejiang, China
2
School of Computer Science and Technology, Zhejiang Normal University, Yingbin Avenue, 321004, Jinhua, Zhejiang, China
3
School of Public Health, Chongqing Medical University, 400016, Chongqing, China
Received:
2
November
2023
Accepted:
2
March
2024
Published online:
14
March
2024
Higher-order relationships are prevalent in the real world, and the strength differences between these relationships are often significant and cannot be ignored. Weighted hypernetworks can more accurately represent these prevalent relationships than simple graphs. Due to various limitations, we may not be able to observe all Higher-order relationships that exist within the network. Therefore, algorithms that can perform weighted hyperlink prediction are of great significance. However, current research on link prediction often focuses on simple networks or ordinary higher-order relationships, without considering the important factor of weight. This paper proposes an algorithm based on topologic similarity to predict whether weighted hyperedges are included in weighted hypernetworks, complementing research in this area. On the artificial weighted hypernetwork, the co-author network of various subjects, and the patent Chinese medicine prescription network, we tested the algorithm’s robustness and achieved higher accuracy than other methods. Applying this algorithm can handle the prediction of weighted many-to-many relationships and improve accuracy by utilizing the weights on the weighted hypernetwork. Compared to the unweighted version, the algorithm’s accuracy on the patent Chinese medicine prescription network has been improved from 97 to 99%. It has been improved from 94 to 96% on the geology co-author network.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2024. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
