# US 2023/0274195 A1 · Time-varying features via metadata

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0274195 A1
- Document ID: US20230274195A1
- Status: Published application
- Kind: A1
- Application number: 17/854,482
- Filed: 2022-06-30
- Published: 2023-08-31
- Earliest claimed: 2022-02-28
- Assignees: Oracle International Corporation
- Applicants: Oracle International Corporation
- Inventors: Ankit Kumar Aggarwal; Vikas Pandey; Praneet Pabolu; Samik Raychaudhuri
- Web page: https://patentgrep.com/patent/US20230274195A1

## CPC

- G06N 20/20: Ensemble learning
- G06N 3/08: Learning methods
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

The present embodiments relate to using feature engineering to generate time-varying features via metadata. A first exemplary embodiment provides a method for performing feature engineering to generate time-varying features. The method can include receiving a first value and a second value of the time-series data. The method can further include receiving metadata that describes a relationship between the first value and the second value. The method can further include detecting the relationship between the first value and the second value based on the metadata. The method can further include generating, a time-varying feature from a combination of the first value and the second value based on the relationship detected from the metadata. The method can further include generating, by implementing the machine learning forecasting model, a forecasted value for the time-series data based on the time-varying feature.

## Claims

### Claim 1

A computer-implemented method, the method comprising:
receiving, by a computing device, a first value associated with a first time step of time-series data and a second value associated with a second timestep of the time-series data;
receiving, by the computing device, metadata that describes a relationship between the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data, the metadata being generated based at least in part on the time-series data;
detecting, by the computing device, the relationship between the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data based at least in part on the metadata;
generating, by the computing device, a time-varying feature from a combination of the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data based at least in part on the relationship detected from the metadata;
receiving, by the computing device, an exogenous data value, the exogenous data value being generated distinctly from the time-series data;
generating, by the computing device, an input data value for a machine learning forecasting model by applying the exogenous data value to the time-varying feature; and
generating, by the computing device implementing the machine learning forecasting model, a forecasted value for the time-series data based at least in part on the input data value.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, further comprising transmitting the input data value to the machine learning forecasting model.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, further comprising outputting a local explanation, a global explanation, a fitted series, and a rolling-origin cross-validation error.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, further comprising adding together the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein generating the time-varying feature further comprises:
selecting a lag value;
identifying the second value associated with the second time step of the time-series data based at least in part on the selected lag value; and
combining the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein the machine learning forecasting model implements a gradient boosting technique to generate the forecasted value for the time-series data.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein the time-varying feature comprises a multi-time step rolling mean value.

### Claim 8

A computing system, comprising:
a processor; and
a computer-readable medium including instructions that, when executed by the processor, cause the processor to:
receive a first value associated with a first time step of time-series data and a second value associated with a second timestep of the time-series data;
receive metadata that describes a relationship between the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data, the metadata being generated based at least in part on the time-series data;
detect the relationship between the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data based at least in part on the metadata;
generate a time-varying feature from a combination of the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data based at least in part on the relationship detected from the metadata;
receive an exogenous data value, the exogenous data value being generated distinctly from the time-series data;
generate an input data value for a machine learning forecasting model by applying the exogenous data value to the time-varying feature; and
generate, by implementing the machine learning forecasting model, a forecasted value for the time-series data based at least in part on the input data value.

### Claim 9 (depends on 8)

The computing system of claim 8, wherein the processor further transmits the input data value to the machine learning forecasting model.

### Claim 10 (depends on 8)

The computing system of claim 8, wherein the processor further outputs a local explanation, a global explanation, a fitted series, and a rolling-origin cross-validation error.

### Claim 11 (depends on 8)

The cloud infrastructure node of claim 8, wherein generating the time-varying feature comprises adding together the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data.

### Claim 12 (depends on 8)

The computing system of claim 8, wherein generating the time-varying feature further comprises:
selecting a lag value;
identifying the second value associated with the second time step of the time-series data based at least in part on the selected lag value; and
combining the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data.

### Claim 13 (depends on 8)

The computing system of claim 8, wherein the machine learning forecasting model implements a gradient boosting technique to generate the forecasted value for the time-series data.

### Claim 14 (depends on 8)

The computing system of claim 8, wherein the time-varying feature comprises a multi-time step rolling mean value.

### Claim 15

A non-transitory computer-readable medium having stored thereon a sequence of instructions which, when executed by a processor, causes the processor to perform operations comprising:
receiving a first value associated with a first time step of time-series data and a second value associated with a second timestep of the time-series data;
receiving metadata that describes a relationship between the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data, the metadata being generated based at least in part on the time-series data;
detecting the relationship between the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data based at least in part on the metadata;
generating a time-varying feature from a combination of the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data based at least in part on the relationship detected from the metadata;
receiving an exogenous data value, the exogenous data value being generated distinctly from the time-series data;
generating an input data value for a machine learning forecasting model by applying the exogenous data value to the time-varying feature; and
generating, by implementing the machine learning forecasting model, a forecasted value for the time-series data based at least in part on the input data value.

### Claim 16 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the operations further comprise transmitting the input data value to the machine learning forecasting model.

### Claim 17 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the operations further comprise outputting a forecasted value based at least in part on the input data, the forecasted value being forecasted from the time-series data.

### Claim 18 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein
generating the time-varying feature comprises adding together the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data.

### Claim 19 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein generating the time-varying feature further comprises:
selecting a lag value;
identifying the second value associated with the second time step of the time-series data based at least in part on the selected lag value; and
combining the first value associated with the first time step of the time-series data and the second value associated with the second time step of the time-series data.

### Claim 20 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the machine learning forecasting model implements a gradient boosting technique to generate the forecasted value for the time-series data.
