摘要:Retraction of:
Scientific Reports
https://doi.org/10.1038/s41598-021-87523-1, published online 15 April 2021
The Editors have retracted this Article.
The testing accuracy for the method reported in the Article was found to be over 99% for all datasets. However, it has been brought to the Editors' attention after the publication of the Article that the graph labels were included as node-level attributes. These attributes were then used by the classifier to predict the graph label, violating the testing procedure, and leading to an over-estimation of the model's classification accuracy.
Ali Ahmadian and Massimiliano Ferrara disagree with this retraction. Pritam Saha, Debadyuti Mukherjee, Pawan Kumar Singh and Ram Sarkar have not responded to correspondence from the Editors about this retraction.