C3.ai has been granted a patent for a computer-implemented method that applies machine learning to detect and attribute network interruptions to specific components or nodes within the network. The method involves mapping a network with dynamically changing islands and using a disaggregation model to detect and localize events at both the individual component and island levels. The patent also includes a method for training a machine learning model to identify the source individual component causing an island-level event. GlobalData’s report on C3.ai gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
According to GlobalData’s company profile on C3.ai, AI for workflow management was a key innovation area identified from patents. C3.ai's grant share as of September 2023 was 22%. Grant share is based on the ratio of number of grants to total number of patents.
Machine learning model for detecting network interruptions and their sources
A recently granted patent (Publication Number: US11777813B2) describes a computer-implemented method for training a machine learning model to identify the source of an island-level event in a dynamic network. The network consists of multiple islands that can change by splitting or merging with other islands. The method involves receiving island-level event data that does not immediately identify the source individual component of the event. The method then estimates the prior probability of each individual component causing the event and performs a posterior inference using a machine learning model to determine the probability estimate of each component causing the event. The prior probability is updated based on the result of the posterior inference, and this process is repeated until the difference between the updated and estimated prior probabilities of each component is below a threshold difference. The machine learning model is trained to receive the island-level event data and identify the source individual component as the island dynamically changes.
The patent also describes a system for training the machine learning model, which includes a processor and a memory device. The memory device contains instructions that configure the processor to perform the operations described in the method. Additionally, a non-transitory computer-readable medium is mentioned, which stores a program that causes a computer to perform the same operations as the method.
The patent further includes various claims related to the method and system. These claims specify additional details such as the use of an exact posterior inference, the initial known or estimated probability of each individual component causing the event based on previous event data, the iterative updating of prior probabilities using an Expectation Maximization (EM) algorithm, the use of an electrical distribution network with nodes and branches as the network, the collection of island-level event data from a subset of individual components, and the use of a machine learning latent variable model including a Bayesian model or a mixture model as the machine learning model.
Overall, this patent presents a method and system for training a machine learning model to identify the source of island-level events in a dynamic network. The method involves estimating prior probabilities, performing posterior inference, and updating the prior probabilities iteratively until a threshold difference is reached. The system includes a processor and memory device, while the non-transitory computer-readable medium stores a program for performing the method. The patent claims provide additional details and variations of the method and system.
To know more about GlobalData’s detailed insights on C3.ai, buy the report here.
Premium Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
