US 12,455,779 B2
Processing natural language network queries
Krishna Prabhakar, Los Altos Hills, CA (US); Rajarao Bhagya Prasad Nittur, Milpitas, CA (US); Anoop Kumaran Nair, Thrikkakara (IN); Rajesh Kumar Ganapathy Achari, Perumbavoor (IN); Madhusudhan Chitradurga Sethuram Setty, Bengaluru (IN); and Venkatesh Ramachandran, Bengaluru (IN)
Assigned to Arista Networks, Inc., Santa Clara, CA (US)
Filed by Arista Networks, Inc., Santa Clara, CA (US)
Filed on Oct. 12, 2023, as Appl. No. 18/486,051.
Claims priority of application No. 202341039926 (IN), filed on Jun. 12, 2023.
Prior Publication US 2024/0411752 A1, Dec. 12, 2024
Int. Cl. G06F 11/07 (2006.01); G06F 11/22 (2006.01); G06F 16/242 (2019.01); G06F 16/245 (2019.01); G06F 40/295 (2020.01); G06F 40/40 (2020.01); G06N 5/022 (2023.01); H04L 9/08 (2006.01); H04L 9/40 (2022.01); H04L 41/0806 (2022.01); H04L 41/0893 (2022.01); H04L 41/0894 (2022.01); H04L 41/0895 (2022.01); H04L 41/14 (2022.01); H04L 41/149 (2022.01); H04L 41/28 (2022.01); H04L 43/0876 (2022.01)
CPC G06F 11/0709 (2013.01) [G06F 11/079 (2013.01); G06F 11/0793 (2013.01); G06F 11/2257 (2013.01); G06F 16/242 (2019.01); G06F 16/245 (2019.01); G06F 40/295 (2020.01); G06F 40/40 (2020.01); G06N 5/022 (2013.01); H04L 9/0825 (2013.01); H04L 41/0806 (2013.01); H04L 41/0893 (2013.01); H04L 41/0894 (2022.05); H04L 41/0895 (2022.05); H04L 41/145 (2013.01); H04L 41/149 (2022.05); H04L 41/28 (2013.01); H04L 43/0876 (2013.01); H04L 63/0823 (2013.01); H04L 63/083 (2013.01); H04L 63/0876 (2013.01); H04L 63/10 (2013.01); H04L 63/102 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method of operating a network associated with a tenant, the method comprising:
receiving a network-related query;
determining, based on a first language model, an intent of the network-related query;
communicating with one or more network devices in the network to obtain network records associated with the one or more network devices for training a second language model, different than the first language model, based on the network records;
extracting, based on the second language model, a named entity from the network-related query by extracting one or more of: a user, a media access control (MAC) address, an Internet Protocol (IP) address, a device type, a location, a network domain, a network service, and a network segment;
storing, on a multi-tenant database, a plurality of named entity lists associated with a plurality of tenants; and
obtaining tenant-specific network-related search results by searching for the extracted named entity on a named entity list corresponding to the tenant in the plurality of named entity lists stored on the multi-tenant database.