Due to the large size of the dataset, arXiv data (graph-dataset and preprocessed textual-dataset) will be shared separately upon paper acceptance on a public repository. 
