Abstract
Device tagging is an important element in the world of network administration, offering an efficient way to organize network and computation resources (such as, for example, network devices, virtual machines, instances, etc.) and support efficient device provisioning and network segmentation (e.g., firewall rules, routing rules, etc.). The manual selection of tags and labelling of individual devices may be error prone and quite time consuming, particularly as the scale of a network grows. To address such challenges, techniques are presented herein that leverage aspects of Graph Convolutional Network (GCN) theory to offer a GCN-based approach for the accurate and automatic tagging of network devices employing a semi-supervised deep learning approach and requiring only minimal human expert knowledge (e.g., for training).
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Zhang, Xinjun; Zacks, Dave; and Shao, Qihong, "SEMI-SUPERVISED DEVICE TAG PREDICTION FOR AUTOMATIC NETWORK PROVISIONING", Technical Disclosure Commons, (September 29, 2020)
https://www.tdcommons.org/dpubs_series/3636