Prediction of higher-order links using global vectors and Hasse diagrams
Publication Date
1-1-2021
Document Type
Conference Proceeding
Publication Title
Proceedings - 2021 IEEE International Conference on Big Data, Big Data 2021
DOI
10.1109/BigData52589.2021.9671432
First Page
4802
Last Page
4811
Abstract
The primary objective of this work is to utilize the GloVeNoR node embedding technique, as well as the Simplex2Vec triangle embedding technique, to perform higher-order link prediction, i.e., the possibility of an interaction of more than two nodes. Additionally, we evaluate the predictions generated by our methods and compare them with existing higher-order link prediction approaches using various benchmark datasets. Based on our experiments, we show that the triangle embeddings generated using our techniques increase the average performance over the five datasets evaluated using the AUC-PR relative to random baseline as a metric for higher-order link prediction.
Keywords
global vectors, graph/node embeddings, Hasse diagram, higher-order structures, link prediction
Department
Computer Science
Recommended Citation
Kalpnil Anjan, Willam Andreopoulos, and Katerina Potika. "Prediction of higher-order links using global vectors and Hasse diagrams" Proceedings - 2021 IEEE International Conference on Big Data, Big Data 2021 (2021): 4802-4811. https://doi.org/10.1109/BigData52589.2021.9671432
