Fortinet has patented a method for training a machine learning model using intelligently selected multiclass vectors. The process involves translating, clustering, and labeling feature vectors to optimize cluster selection and improve classification accuracy for computing devices. GlobalData’s report on Fortinet gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
According to GlobalData’s company profile on Fortinet, was a key innovation area identified from patents. Fortinet's grant share as of May 2024 was 47%. Grant share is based on the ratio of number of grants to total number of patents.
Machine learning model training using selected multiclass vectors
A recently granted patent (Publication Number: US12001515B2) outlines a method and system for machine learning model training and classification of computing devices. The method involves receiving a set of un-labeled feature vectors, translating them using a T-Distributed Stochastic Neighbor Embedding (t-SNE) process to reduce dimensionality, clustering the vectors, and identifying optimal clusters through convex optimization. Representative vectors are selected from these clusters for labeling, creating a set of labeled feature vectors based on received labels, and training a machine learning model for multi-class classification. The model is then executed to classify computing devices based on the initial set of feature vectors.
The patent also details the system components, including processing circuitry and a non-transitory computer-readable medium storing instructions for executing the method. The system follows a similar process of receiving, translating, clustering, identifying optimal clusters, selecting representative vectors, labeling, training the model, and executing it for classification. Notably, the method involves selecting boundary condition feature vectors for labeling by an oracle based on prediction skepticism scores, enhancing the accuracy of the classification process. The system is designed to handle a variety of computing devices, particularly emphasizing network security applications, and can process feature vectors containing Gaussian noise data. Overall, the patent presents a comprehensive approach to machine learning model training and classification, particularly suited for multi-class classification tasks in the realm of computing devices.
To know more about GlobalData’s detailed insights on Fortinet, buy the report here.
Data 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.
