US 11,983,186 B2
Predicting potential incident event data structures based on multi-modal analysis
Bhabesh Acharya, Bangalore (IN); Prajosh Thaykandy, Calicut (IN); Raja Sekhar Kommoju, Bangalore (IN); and Raja Pratap Kondamari, Bangalore (IN)
Assigned to HONEYWELL INTERNATIONAL INC., Charlotte, NC (US)
Filed by Honeywell International Inc., Charlotte, NC (US)
Filed on Oct. 6, 2020, as Appl. No. 17/064,337.
Application 17/064,337 is a continuation in part of application No. 16/661,884, filed on Oct. 23, 2019.
Prior Publication US 2021/0124741 A1, Apr. 29, 2021
Int. Cl. G06F 16/2458 (2019.01); G06F 16/22 (2019.01); G06N 5/02 (2023.01); G06N 20/00 (2019.01); G06Q 50/26 (2012.01); G06V 20/40 (2022.01)
CPC G06F 16/2465 (2019.01) [G06F 16/2228 (2019.01); G06N 5/02 (2013.01); G06Q 50/26 (2013.01); G06V 20/41 (2022.01); G06N 20/00 (2019.01); G06V 20/44 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
at a device with one or more processors and a memory:
receiving a request comprising incident data indicative of at least one of: a time and a location of occurrence of an incident, wherein the request is based on an occurrence of the incident;
transmitting the request to a plurality of third party resource providers configured to provide first information associated with the incident, wherein the plurality of third party resource providers are disparate database sources, and wherein the plurality of third party resource providers are selected based at least in part on the incident data;
processing, by a trained machine learning model, the first information received from the plurality of third party resource providers to identify one or more incident descriptions;
identifying, by the trained machine learning model, one or more potential incident descriptions from amongst the one or more incident descriptions;
generating, by the trained machine learning model, an ontology structure for the one or more potential incident descriptions, the ontology structure indicating a relationship between the one or more potential incident descriptions;
analyzing one or more sensors present in the location of occurrence of the incident;
determining, by the trained machine learning model based on the relationship between the one or more potential incident descriptions, at least some of the one or more sensors proximate to the location, wherein at least some of the one or more sensors are configured to provide second information relevant to the incident;
comparing the second information from at least some of the one or more sensors with the ontology structure; and
identifying, by the trained machine learning model, one or more potential persons responsible for the incident in response to comparing the second information with the ontology structure.