Lin Zhao

US Patent:

20120271631, Oct 25, 2012

Filed:

Apr 19, 2012

Appl. No.:

13/450861

Inventors:

Fuliang Weng - Mountain View CA, US
Zhe Feng - Mountain View CA, US
Kui Xu - Sunnyvale CA, US
Lin Zhao - Sunnyvale CA, US

Assignee:

ROBERT BOSCH GMBH - Stuttgart

International Classification:

G10L 15/06

US Classification:

704243, 704E15007

Abstract:

In accordance with one embodiment, a method of generating language models for speech recognition includes identifying a plurality of utterances in training data corresponding to speech, generating a frequency count of each utterance in the plurality of utterances, generating a high-frequency plurality of utterances from the plurality of utterances having a frequency that exceeds a predetermined frequency threshold, generating a low-frequency plurality of utterances from the plurality of utterances having a frequency that is below the predetermined frequency threshold, generating a grammar-based language model using the high-frequency plurality of utterances as training data, and generating a statistical language model using the low-frequency plurality of utterances as training data.

US Patent:

20200342168, Oct 29, 2020

Filed:

Dec 27, 2018

Appl. No.:

16/958892

Inventors:

- Stuttgart, DE
Lin Zhao - Sunnyvale CA, US
Kui Xu - Sunnyvale CA, US
Zhe Feng - Mountain View CA, US

International Classification:

G06F 40/20
G06N 3/04
G06N 3/08

Abstract:

A method of automatically generating a terminology definition knowledge base (KB) includes mapping each word in a word sequence to a real value dense vector using dense vector representations. The word sequence is then processed using a Convolutional Neural Network (CNN) model to identify whether the word sequence includes a terminology definition and to label the word sequence with a label indicating whether a terminology definition exists within the word sequence. The word sequence is then processed using a Conditional Random Field (CRF) model to identify boundaries of the terminology definition in the word sequence. The terminology definition is then extracted and added to the terminology definition KB.

US Patent:

20200342178, Oct 29, 2020

Filed:

Dec 27, 2018

Appl. No.:

16/958888

Inventors:

- Stuttgart, DE
Yifan He - Sunnyvale CA, US
Lin Zhao - Sunnyvale CA, US
Kui Xu - Sunnyvale CA, US
Zhe Feng - Mountain View CA, US

International Classification:

G06F 40/30
G06N 5/02
G06F 16/36
G06N 5/04
G06F 40/169
G06F 40/247
G06N 7/00

Abstract:

An automatic terminology linking system includes a candidate generator configured to identify candidate nodes for each terminology that is to be linked to a node of the knowledge base. A pseudo-candidate generator is configured to identify pseudo-candidate nodes for candidate-less terminologies. A candidate scorer is configured to respectively score the candidate nodes and the pseudo-candidate nodes by collective inference using occurrence statistics and co-occurrence statistics for these nodes. The pseudo-candidate generator is configured to identify knowledge base nodes that are semantically-related to candidate-less terminology as the pseudo-candidate nodes for the candidate-less terminology.

US Patent:

20200304448, Sep 24, 2020

Filed:

Dec 1, 2016

Appl. No.:

15/779592

Inventors:

- Stuttgart, DE
Lin Zhao - Sunnyvale CA, US
Kui Xu - Sunnyvale CA, US

International Classification:

H04L 12/58
G06F 40/279
G06F 40/258

Abstract:

A non-transitory machine accessible medium includes a topic detection module for extracting major topics from an electronic mail message, a sender role and receiver role detection module for detecting role information of one of the sender and the receiver from the electronic mail message, and a relationship detection module for comparing a relationship between the extracted topic information and the role information. The detection modules can be based on statistical or rule-based approaches. Based on the detected information, the non-transitory machine accessible medium predicts a level of importance of the electronic mail message based on a machine learning algorithm.

US Patent:

20200210426, Jul 2, 2020

Filed:

Dec 28, 2018

Appl. No.:

16/234989

Inventors:

- Stuttgart, DE
Lin ZHAO - Sunnyvale CA, US
Zhe FENG - Mountain View CA, US

International Classification:

G06F 16/2452
G06F 16/9538
G06F 16/248

Abstract:

A server includes a controller in communication with the server and configured to receive a query related to an issue with a vehicle, retrieve information related to the query from a database that includes structured data including data related to vehicle issues and a solution to the vehicle issues, determine whether the database includes structured data including data related to the solution responsive to the query, and output information including the solution to the issue when the determination identifies that the structured data including the solution in the database or retrieve information from a website when the determination identifies that the structured data including the solution is not found in the database, wherein the controller is further configured to retrieve the information from the website by conducting a keyword search for a first set of results and filtering the information from the website utilizing an autoencoder.

US Patent:

20190244788, Aug 8, 2019

Filed:

Apr 15, 2019

Appl. No.:

16/384442

Inventors:

- Fremont CA, US
Felix Kozakevich - Sunnyvale CA, US
Michael C. Kellogg - San Francisco CA, US
John Patrick Holland - San Jose CA, US
Zhigang Chen - Campbell CA, US
Kenneth Lucchesi - Newark CA, US
Lin Zhao - Fremont CA, US

International Classification:

H01J 37/32
H01L 21/3065
H01L 21/67
H01L 21/687

Abstract:

Systems and methods for achieving a pre-determined factor associated with the edge region within the plasma chamber is described. One of the methods includes providing an RF signal to a main electrode within the plasma chamber. The RF signal is generated based on a frequency of operation of a first RF generator. The method further includes providing another RF signal to an edge electrode within the plasma chamber. The other RF signal is generated based on the frequency of operation of the first RF generator. The method includes receiving a first measurement of a variable, receiving a second measurement of the variable, and modifying a phase of the other RF signal based on the first measurement and the second measurement. The method includes changing a magnitude of a variable associated with a second RF generator to achieve the pre-determined factor.

US Patent:

20190080885, Mar 14, 2019

Filed:

Jun 14, 2018

Appl. No.:

16/008529

Inventors:

- Fremont CA, US
Lin Zhao - Fremont CA, US
Felix Kozakevich - Sunnyvale CA, US
Kenneth Lucchesi - Newark CA, US
Zhigang Chen - Campbell CA, US
John Patrick Holland - San Jose CA, US

International Classification:

H01J 37/32

Abstract:

Systems and methods for providing a multi regime plasma wafer processing are described. The systems and methods have three states. During a first one of the states, an etching operation is performed. In a second one of the states, a power level of a kilohertz radio frequency signal is greater than zero to increase directionality of ions that are incident on a bottom surface of a stack layer. In a third one of the states, there is a reduction in a loss of mask on top of the stack layer and deposition may be performed.

US Patent:

20190005027, Jan 3, 2019

Filed:

Jun 22, 2018

Appl. No.:

16/016106

Inventors:

- Stuttgart, DE
Lin Zhao - Sunnyvale CA, US
Kui Xu - Sunnyvale CA, US
Zhe Feng - Mountain View CA, US

International Classification:

G06F 17/27
G06N 99/00
G06Q 50/00
G06Q 30/02
G06F 17/28

Abstract:

A method for automated aspect-based sentiment analysis includes parsing reviews from a first domain to generate rhetorical structure trees and extracting rhetorical rules from the rhetorical structure trees, each rhetorical rule including a path extracted from at least one span in at least one of the rhetorical structure trees associated with a probability that the path corresponds to a positive or negative sentiment based on annotation data. The method further includes parsing reviews from a second domain to generate a second plurality of rhetorical structure trees, generating training data that associates at least one aspect in the review from the second domain with a sentiment associated with a rhetorical rule in the plurality of rhetorical rules, and training a classifier to identify sentiments in reviews from the second domain using the second plurality of reviews and the training data.