-
公开(公告)号:US20210182912A1
公开(公告)日:2021-06-17
申请号:US16866798
申请日:2020-05-05
发明人: Sanjeev Misra , Appavu Siva Prakasam , Ann Eileen Skudlark , Siva Kolachina , Nisha Shahul Hameed , Prashanth Boddhireddy , Lien Tran , Jenq-Chyuan Wang
摘要: Aspects of the subject disclosure may include, for example, determining classes from a corpus based on topic modeling, data clustering and unsupervised learning. Labels are determined for each of the classes and trained models are generated for each of the classes by assignment of a plurality of textual documents to labels based on a highest number of matches. A raw textual document can be tokenized and stop words removed. A corresponding one of the trained models can be selected according to a class that is applicable to subject matter of the raw textual document. The processed document can be assigned to a target label based on a highest number of matches of words. Other embodiments are disclosed.
-
公开(公告)号:US11386463B2
公开(公告)日:2022-07-12
申请号:US16866798
申请日:2020-05-05
发明人: Sanjeev Misra , Appavu Siva Prakasam , Ann Eileen Skudlark , Siva Kolachina , Nisha Shahul Hameed , Prashanth Boddhireddy , Lien Tran , Jenq-Chyuan Wang
摘要: Aspects of the subject disclosure may include, for example, determining classes from a corpus based on topic modeling, data clustering and unsupervised learning. Labels are determined for each of the classes and trained models are generated for each of the classes by assignment of a plurality of textual documents to labels based on a highest number of matches. A raw textual document can be tokenized and stop words removed. A corresponding one of the trained models can be selected according to a class that is applicable to subject matter of the raw textual document. The processed document can be assigned to a target label based on a highest number of matches of words. Other embodiments are disclosed.
-
公开(公告)号:US11336743B2
公开(公告)日:2022-05-17
申请号:US17201438
申请日:2021-03-15
发明人: Ann E. Skudlark , Eric Zavesky , Lien Tran , David Crawford Gibbon , Zhu Liu
IPC分类号: H04L29/08 , H04N21/25 , H04L67/306 , H04L67/50
摘要: Aspects of the subject disclosure may include, for example, a method that includes obtaining metadata from media content and consumed by network subscribers; determining for each network subscriber a consumer context associated with the media content; and determining a media consumption pattern for each network subscriber based on the metadata and the consumer context, thereby generating a plurality of media consumption patterns. The method further includes aggregating the media consumption patterns; determining, based on the aggregated media consumption patterns, a media consumption trend for the network subscribers; and correlating the media consumption trend with a profile including a current activity for a network subscriber of the plurality of network subscribers, thereby generating a recommendation for the network subscriber regarding new media content not previously consumed by the network subscriber. The method also includes communicating the recommendation to the network subscriber. Other embodiments are disclosed.
-
公开(公告)号:US10979528B2
公开(公告)日:2021-04-13
申请号:US16222323
申请日:2018-12-17
发明人: Ann E. Skudlark , Eric Zavesky , Lien Tran , David Crawford Gibbon , Zhu Liu
摘要: Aspects of the subject disclosure may include, for example, a method that includes obtaining metadata from media content and consumed by network subscribers; determining for each network subscriber a consumer context associated with the media content; and determining a media consumption pattern for each network subscriber based on the metadata and the consumer context, thereby generating a plurality of media consumption patterns. The method further includes aggregating the media consumption patterns; determining, based on the aggregated media consumption patterns, a media consumption trend for the network subscribers; and correlating the media consumption trend with a profile including a current activity for a network subscriber of the plurality of network subscribers, thereby generating a recommendation for the network subscriber regarding new media content not previously consumed by the network subscriber. The method also includes communicating the recommendation to the network subscriber. Other embodiments are disclosed.
-
公开(公告)号:US10958782B2
公开(公告)日:2021-03-23
申请号:US16434277
申请日:2019-06-07
发明人: Lien Tran , Simon Byers , Carlos Eduardo De Andrade , David Poole , Emir Halepovic , Vijay Gopalakrishnan , Christopher Volinsky
摘要: Software downloads to Internet of things (IoT) devices are facilitated over a constrained network. In one embodiment a method comprises monitoring, by a network device comprising a processor, data determined to have been sent to a device for transmission to the device via a radio access network device of a wireless communication network, and determining, by the network device, a type of traffic associated with the data. The method further includes based on a determination that the data comprises firmware and that the type of traffic is of a traffic priority that is lower than a defined traffic priority, applying, by the network device, a low priority transport protocol to the data, wherein the applying comprises associating protocol information with the data representative of the low priority transport protocol.
-
公开(公告)号:US20220224768A1
公开(公告)日:2022-07-14
申请号:US17712975
申请日:2022-04-04
发明人: Ann E. Skudlark , Eric Zavesky , Lien Tran , David Crawford Gibbon , Zhu Liu
IPC分类号: H04L67/306 , H04N21/25 , H04L67/50 , H04N21/442
摘要: Aspects of the subject disclosure may include, for example, a method that includes obtaining metadata from media content and consumed by network subscribers; determining for each network subscriber a consumer context associated with the media content; and determining a media consumption pattern for each network subscriber based on the metadata and the consumer context, thereby generating a plurality of media consumption patterns. The method further includes aggregating the media consumption patterns; determining, based on the aggregated media consumption patterns, a media consumption trend for the network subscribers; and correlating the media consumption trend with a profile including a current activity for a network subscriber of the plurality of network subscribers, thereby generating a recommendation for the network subscriber regarding new media content not previously consumed by the network subscriber. The method also includes communicating the recommendation to the network subscriber. Other embodiments are disclosed.
-
公开(公告)号:US20200022147A1
公开(公告)日:2020-01-16
申请号:US16584102
申请日:2019-09-26
发明人: Lien Tran , Simon D. Byers , Carlos Eduardo De Andrade , Emir Halepovic , David John Poole , Christopher T. Volinsky
摘要: A more efficient over-the-air software push can be facilitated by leveraging a smart scheduling system for vehicles. The smart scheduling system can use location and network capacity data to prioritize over-the-air software pushes for vehicles. For instance, a vehicle, which is only operational during off-peak wireless network hours can receive a software push during the off-peak times because wireless network capacity is not an issue. However, vehicles, which are used primarily during heavy peak wireless network times can receive software in a prioritized manner based on location data, frequency of use, network capacity, etc.
-
公开(公告)号:US10362166B2
公开(公告)日:2019-07-23
申请号:US15446797
申请日:2017-03-01
发明人: Lien Tran , Simon Byers , Carlos Eduardo De Andrade , David Poole , Emir Halepovic , Vijay Gopalakrishnan , Christopher Volinsky
摘要: Software downloads to Internet of things (IoT) devices are facilitated over a constrained network. In one embodiment a method comprises monitoring, by a network device comprising a processor, data determined to have been sent to a device for transmission to the device via a radio access network device of a wireless communication network, and determining, by the network device, a type of traffic associated with the data. The method further includes based on a determination that the data comprises firmware and that the type of traffic is of a traffic priority that is lower than a defined traffic priority, applying, by the network device, a low priority transport protocol to the data, wherein the applying comprises associating protocol information with the data representative of the low priority transport protocol.
-
公开(公告)号:US20220277351A1
公开(公告)日:2022-09-01
申请号:US17749347
申请日:2022-05-20
发明人: Sanjeev Misra , Appavu Siva Prakasam , Ann Eileen Skudlark , Siva Kolachina , Nisha Shahul Hameed , Prashanth Boddhireddy , Lien Tran , Jenq-Chyuan Wang
摘要: Aspects of the subject disclosure may include, for example, determining classes from a corpus based on topic modeling, data clustering and unsupervised learning. Labels are determined for each of the classes and trained models are generated for each of the classes by assignment of a plurality of textual documents to labels based on a highest number of matches. A raw textual document can be tokenized and stop words removed. A corresponding one of the trained models can be selected according to a class that is applicable to subject matter of the raw textual document. The processed document can be assigned to a target label based on a highest number of matches of words. Other embodiments are disclosed.
-
公开(公告)号:US20210203749A1
公开(公告)日:2021-07-01
申请号:US17201438
申请日:2021-03-15
发明人: Ann E. Skudlark , Eric Zavesky , Lien Tran , David Crawford Gibbon , Zhu Liu
摘要: Aspects of the subject disclosure may include, for example, a method that includes obtaining metadata from media content and consumed by network subscribers; determining for each network subscriber a consumer context associated with the media content; and determining a media consumption pattern for each network subscriber based on the metadata and the consumer context, thereby generating a plurality of media consumption patterns. The method further includes aggregating the media consumption patterns; determining, based on the aggregated media consumption patterns, a media consumption trend for the network subscribers; and correlating the media consumption trend with a profile including a current activity for a network subscriber of the plurality of network subscribers, thereby generating a recommendation for the network subscriber regarding new media content not previously consumed by the network subscriber. The method also includes communicating the recommendation to the network subscriber. Other embodiments are disclosed.
-
-
-
-
-
-
-
-
-