Yee M Chan

US Patent:

20220180073, Jun 9, 2022

Filed:

Dec 16, 2021

Appl. No.:

17/553312

Inventors:

- Cambridge MA, US
Yee Seng Chan - Winchester MA, US
Ilana Heintz - Silver Spring MD, US

International Classification:

G06F 40/58
G06N 20/00
G06F 40/253
G06F 40/30
G06N 3/04

Abstract:

A machine accesses a preexisting set of natural language text documents in multiple natural languages. Each natural language text document in at least a portion of the preexisting set is associated with an event. The machine trains, using the preexisting set of natural language text documents and the associated events, an event encoder to learn associations between texts and event annotations. The event encoder leverages a parser in each of the two or more natural languages. The machine generates, using the event encoder, new event annotations for texts. The machine trains, using the preexisting set of natural language text documents and the new event annotations for the texts generated by the event encoder, an event extraction engine to extract events from natural language texts in the two or more natural languages. The event extraction engine leverages the parser in each of the two or more natural languages.

US Patent:

20200387574, Dec 10, 2020

Filed:

Jun 7, 2019

Appl. No.:

16/434710

Inventors:

- Cambridge MA, US
Yee Seng Chan - Winchester MA, US
Ilana Heintz - Silver Spring MD, US

International Classification:

G06F 17/28
G06F 17/27
G06N 3/04
G06N 20/00

Abstract:

A machine accesses a preexisting set of natural language text documents in multiple natural languages. Each natural language text document in at least a portion of the preexisting set is associated with an event. The machine trains, using the preexisting set of natural language text documents and the associated events, an event encoder to learn associations between texts and event annotations. The event encoder leverages a parser in each of the two or more natural languages. The machine generates, using the event encoder, new event annotations for texts. The machine trains, using the preexisting set of natural language text documents and the new event annotations for the texts generated by the event encoder, an event extraction engine to extract events from natural language texts in the two or more natural languages. The event extraction engine leverages the parser in each of the two or more natural languages.

US Patent:

20190335343, Oct 31, 2019

Filed:

Oct 29, 2018

Appl. No.:

16/173713

Inventors:

- Arlington VA, US
Lily Zhu - Parsippany NJ, US
Yee Sin Chan - San Jose CA, US
William H. Stone - Doylestown PA, US
Gregory Dial - Flanders NJ, US
Ahmed Moussa - Basking Ridge NJ, US
Christopher M. Schmidt - Branchburg NJ, US
Michael Kondratiuk - Hillsborough NJ, US
Gerard J. Flynn - Washington NJ, US
Balaji L. Raghavachari - Bridgewater NJ, US

International Classification:

H04W 24/02
H04W 76/10

Abstract:

A user device receives system information block data from a cell. The user device obtains, from the system information block data, cell metadata that includes a cellular parameter. The user device parses the cell metadata to identify the cellular parameter. The cellular parameter includes an upper layer indicator element, a public land mobile network (PLMN) information list element, and a network capability indicator element. The user device determines, based on the cellular parameter, that the cell is of a particular cell type of a plurality of cell types, and that the cell has a particular capability of a plurality of types of capabilities. The user device performs an action based on the particular cell type or the particular capability.

US Patent:

20160044538, Feb 11, 2016

Filed:

Aug 11, 2014

Appl. No.:

14/456756

Inventors:

- Arlington VA, US
Max A. Solondz - New Vernon NJ, US
Yee Sin Chan - San Jose CA, US

International Classification:

H04W 28/08
H04W 36/30

Abstract:

A wireless telecommunications system may be configured to provide wireless service via licensed and/or shared frequency bands (e.g., frequency bands that are shared with other telecommunications systems, and for which access is regulated by an external entity). A load balancing technique, described herein, may provide load balancing between licensed and shared frequency bands based on traffic type (e.g., real-time or non-real-time traffic) and/or based on device type (e.g., whether a device is typically involved in real-time communications). Techniques described herein may also provide for the staggered or gradual reduction of the transmit power of a radio associated with a shared frequency band (e.g., in response to an instruction from the external entity), in order to minimize the abrupt increase in load on other frequency bands that may result from reducing coverage of the shared frequency band.