US 11,922,520 B2
Determining significant events within an agribusiness system
Sushain Pandit, Austin, TX (US); and Krishna Teja Rekapalli, Austin, TX (US)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Nov. 25, 2020, as Appl. No. 17/104,015.
Prior Publication US 2022/0164900 A1, May 26, 2022
Int. Cl. G06Q 50/02 (2012.01); G06N 3/08 (2023.01); G06N 20/00 (2019.01); G06Q 10/08 (2023.01); G06Q 20/38 (2012.01); G06Q 30/0204 (2023.01); G06Q 50/28 (2012.01)
CPC G06Q 50/02 (2013.01) [G06N 3/08 (2013.01); G06N 20/00 (2019.01); G06Q 10/08 (2013.01); G06Q 20/389 (2013.01); G06Q 30/0205 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-based method for automatically identifying significant events for food traceability for use by a distributed agriculture supply chain, comprising:
training a neural network to automatically identify significant events for food traceability, wherein the neural network comprises a plurality of artificial cells interconnected via a plurality of gates, and wherein each gate encodes a strength of a relationship in the connection between an output of one artificial cell and an input of another artificial cell, the training comprising:
receiving training data, wherein the training data comprises labeled event data for a plurality of farms over a plurality of days;
generating, by the neural network, a probability factor representing a chance that an event in the training data will be significant for food traceability; and
comparing the generated probability to the received labels;
in response to the comparison, updating the encoded strength of at least some of the gates;
receiving data from a plurality of sensors, wherein each of the sensors generates a series of events about a respective one or more of a plurality of agriculture supply chain entities;
automatically filtering, by the trained neural network, events in the plurality of series of events having a significance for food traceability less than a threshold; and
automatically selectively transmitting, by a network interface, only the data associated with the unfiltered events to a plurality of blockchain nodes of a distributed ledger.