公开(公告)号：US20240144090A1

公开(公告)日：2024-05-02

申请号：US18385587

申请日：2023-10-31

Applicant: Google LLC

Inventor： Haoming Chen ,  Xi Cheng ,  Weijie Shen ,  Amir Hossein Hormati ,  Honglin Zheng

IPC: G06N20/00

CPC classification number: G06N20/00

Abstract: Aspects of the disclosure are directed to an approach for training a multivariate time series forecasting model using linear regression and ARIMA. The training may be performed by accessing data stored in a data warehouse using structure query language commands. The disclosure further provides for forecasting utilizing the trained model.

公开(公告)号：US20190124377A1

公开(公告)日：2019-04-25

申请号：US15788770

申请日：2017-10-19

Applicant: GOOGLE LLC

Inventor： Haoming Chen ,  Thomas Inskip ,  Kongqun Yang

IPC: H04N21/2389 ,  H04N21/8358 ,  G06F17/30 ,  H04N19/467

CPC classification number: H04N21/23892 ,  G06F16/48 ,  G06F21/10 ,  G06F21/16 ,  H04L9/08 ,  H04L9/16 ,  H04L9/3247 ,  H04N19/467 ,  H04N21/222 ,  H04N21/23424 ,  H04N21/2351 ,  H04N21/23895 ,  H04N21/8358 ,  H04N21/8456 ,  H04N21/85406

Abstract: A method includes generating a first media file that includes a plurality of first media file segments, generating a second media file as a copy of the first media file, the second media file includes a plurality of second media file segments, embedding a first watermark with the plurality of first media file segments, embedding at least one second watermark with the plurality of second media file segments, generating a manifest file based on a portion of the plurality of first media file segments and a portion of the plurality of second media file segments, the manifest file including a plurality of addresses each referencing a media file segment, and encrypting each of plurality of addresses.

公开(公告)号：US20250013937A1

公开(公告)日：2025-01-09

申请号：US18739429

申请日：2024-06-11

Applicant: Google LLC

Inventor： Honglin Zheng ,  Haoming Chen ,  Jun Ya Zhang ,  Xi Cheng ,  Weijie Shen ,  Jiashang Liu ,  Mingge Deng ,  Amir Hossein Hormati

IPC: G06Q10/04 ,  G06Q10/1057

Abstract: Aspects of the disclosure are directed methods, systems, and computer readable media for in-database holiday effect modeling for time series forecasting. The modeling can be accurate, explainable, customizable, and scalable. Machine learning models can receive a first dataset for time series data and a second dataset for configurable holiday data. The models can detect and model effects of each configurable holiday on one or more forecasts, effectively accumulating effects of overlapping holidays, to manage different levels of holiday modeling. Holiday data can be customizable, including an ability to modify existing holidays and/or add new holidays, through one or more interfaces that can display default holiday information, combined holiday information based on both default and customizable holidays, effects of each holiday on forecasts, and accumulated effects of multiple holidays on forecasts.

公开(公告)号：US10432991B2

公开(公告)日：2019-10-01

申请号：US15788770

申请日：2017-10-19

Applicant: GOOGLE LLC

Inventor： Haoming Chen ,  Thomas Inskip ,  Kongqun Yang

IPC: H04N21/2389 ,  H04N19/467 ,  G06F16/48 ,  H04N21/8358 ,  H04N21/222 ,  H04N21/235 ,  H04N21/845 ,  H04N21/854 ,  G06F21/16 ,  H04L9/16 ,  H04L9/08 ,  H04N21/234 ,  G06F21/10 ,  H04L9/32

Abstract: A method includes generating a first media file that includes a plurality of first media file segments, generating a second media file as a copy of the first media file, the second media file includes a plurality of second media file segments, embedding a first watermark with the plurality of first media file segments, embedding at least one second watermark with the plurality of second media file segments, generating a manifest file based on a portion of the plurality of first media file segments and a portion of the plurality of second media file segments, the manifest file including a plurality of addresses each referencing a media file segment, and encrypting each of plurality of addresses.