-
公开(公告)号:US20240414199A1
公开(公告)日:2024-12-12
申请号:US18813106
申请日:2024-08-23
Applicant: AT&T Intellectual Property I, L.P.
Inventor: Antoine Diffloth , Natalie Gilbert , Sundaresan Manoharan
IPC: H04L9/40 , G06F18/2413 , G06N3/04 , H04L61/4511
Abstract: The described technology is generally directed towards homoglyph attack detection. A homoglyph attack detection service can create images of customer's protected domain names. A convolutional neural network can generate feature vectors based on the images. The feature vectors can be stored in a similarity search data store. Newly observed domain names can be compared to the customer's protected domain names, by also generating feature vectors for the newly observed domain names and conducting approximate nearest neighbor searches. Search results can be further evaluated by comparing protected domain names to newly observed domain names using a siamese neural network which applies a similarity threshold. Newly observed domain names that meet or exceed the similarity threshold can be flagged for further action.
-
公开(公告)号:US20220327919A1
公开(公告)日:2022-10-13
申请号:US17227482
申请日:2021-04-12
Applicant: AT&T Intellectual Property I, L.P.
Inventor: Natalie Gilbert , Mazin E. Gilbert
IPC: G08G1/01 , G08G1/0968 , G06N3/08
Abstract: Aspects of the subject disclosure may include, for example, using an AI machine to predict one or more traffic patterns, and the comparing the predicted traffic patterns to actual traffic patterns. Recommended travel paths may be modified as a result. AI machines may be feed-forward or recurrent, and may be trained using any suitable algorithm, including deep learning. Other embodiments are disclosed.
-
公开(公告)号:US12095813B2
公开(公告)日:2024-09-17
申请号:US17380677
申请日:2021-07-20
Applicant: AT&T Intellectual Property I, L.P.
Inventor: Antoine Diffloth , Natalie Gilbert , Sundaresan Manoharan
IPC: H04L9/40 , G06F18/2413 , G06N3/04 , H04L61/4511
CPC classification number: H04L63/1483 , G06F18/24147 , G06N3/04 , H04L61/4511 , H04L63/1416 , H04L63/1425 , H04L63/1475
Abstract: The described technology is generally directed towards homoglyph attack detection. A homoglyph attack detection service can create images of customer's protected domain names. A convolutional neural network can generate feature vectors based on the images. The feature vectors can be stored in a similarity search data store. Newly observed domain names can be compared to the customer's protected domain names, by also generating feature vectors for the newly observed domain names and conducting approximate nearest neighbor searches. Search results can be further evaluated by comparing protected domain names to newly observed domain names using a siamese neural network which applies a similarity threshold. Newly observed domain names that meet or exceed the similarity threshold can be flagged for further action.
-
公开(公告)号:US20230028490A1
公开(公告)日:2023-01-26
申请号:US17380677
申请日:2021-07-20
Applicant: AT&T Intellectual Property I, L.P.
Inventor: Antoine Diffloth , Natalie Gilbert , Sundaresan Manoharan
Abstract: The described technology is generally directed towards homoglyph attack detection. A homoglyph attack detection service can create images of customer's protected domain names. A convolutional neural network can generate feature vectors based on the images. The feature vectors can be stored in a similarity search data store. Newly observed domain names can be compared to the customer's protected domain names, by also generating feature vectors for the newly observed domain names and conducting approximate nearest neighbor searches. Search results can be further evaluated by comparing protected domain names to newly observed domain names using a siamese neural network which applies a similarity threshold. Newly observed domain names that meet or exceed the similarity threshold can be flagged for further action.
-
公开(公告)号:US20210194985A1
公开(公告)日:2021-06-24
申请号:US16723382
申请日:2019-12-20
Applicant: AT&T Intellectual Property I, L.P.
Inventor: Natalie Gilbert , Mazin Gilbert
IPC: H04L29/08 , H04W4/029 , G06K9/00 , G06K9/46 , G10L25/63 , G10L15/22 , G06N20/00 , G06Q30/02 , A61B5/021 , A61B5/00 , A61B5/024 , A61B5/11 , A61B5/16
Abstract: Content presentation is timed based on an individual's predicted mental state. In one example, a method performed by a processing system includes extracting a feature set from data that is collected by at least one sensor, where the sensor is monitoring an individual, the feature set comprises at least one feature of the individual, and the feature set excludes features extracted from images of the individual, predicting a current mental state of the individual, where the current mental state is predicted by providing the feature set as input to a machine learning model, sending media content to an endpoint device of the individual when the current mental state indicates that the individual is likely to be receptive to receiving the media content, and postponing sending media content to the endpoint device when the current mental state indicates that the individual is unlikely to be receptive to receiving the media content.
-
-
-
-