# US 2022/0276997 A1 · Updating metadata associated with historic data

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

- Number: US 2022/0276997 A1
- Document ID: US20220276997A1
- Status: Published application
- Kind: A1
- Application number: 17/189,160
- Filed: 2021-03-01
- Published: 2022-09-01
- Earliest claimed: 2021-03-01
- Assignees: Motorola Solutions, Inc.
- Applicants: Motorola Solutions, Inc.
- Inventors: Zili Li; Xiang Gao; Jari P. Jarvinen
- Web page: https://patentgrep.com/patent/US20220276997A1

## CPC

- G06F 16/2379: {Updates performed during online database operations; commit processing}
- G06F 16/219: {Managing data history or versioning (querying versioned data G06F16/2474; querying temporal data G06F16/2477)}
- G06N 5/02: Knowledge representation; Symbolic representation
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G06N 20/00: Machine learning
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A system for updating metadata associated with historic data. The system includes an electronic computing device. The electronic computing device is configured to receive historic data and metadata associated with the historic data and, using a low capacity machine learning model, analyze the metadata associated with the historic data to determine a probability that the historic data includes the object of interest. The electronic computing device is also configured to compare the probability to a predetermined threshold. The electronic computing device is further configured to use a high capacity machine learning model to analyze the historic data to determine whether the historic data includes the object of interest and update the metadata of the historic data based on whether the historic data includes the object of interest to generate updated metadata, when the probability is greater than or equal to the predetermined threshold.

## Claims

### Claim 1

A system for updating metadata associated with historic data, the system comprising
an electronic computing device, the electronic computing device configured to
receive historic data and metadata associated with the historic data;
using a low capacity machine learning model, analyze the metadata associated with the historic data to determine a probability that the historic data includes an object of interest;
compare the probability to a predetermined threshold; and
when the probability is greater than or equal to the predetermined threshold,
use a high capacity machine learning model to analyze the historic data to determine whether the historic data includes the object of interest; and
update the metadata of the historic data based on whether the historic data includes the object of interest to generate updated metadata.

### Claim 2 (depends on 1)

The system according to claim 1, wherein the electronic computing device is further configured to
when the probability is less than the predetermined threshold, use the low capacity machine learning model to update the metadata of the historic data based on whether the historic data includes the object of interest to generate updated metadata.

### Claim 3 (depends on 1)

The system according to claim 1, wherein the electronic computing device is further configured to
receive new data;
using the high capacity machine learning model, determine metadata associated with the new data, wherein the metadata associated with the new data includes whether or not the new data includes the object of interest; and
use the metadata associated with the new data and the metadata associated with the historic data to train the low capacity machine learning model.

### Claim 4 (depends on 1)

The system according to claim 1, wherein the historic data includes audio data, visual data, textual data, or a combination of the foregoing.

### Claim 5 (depends on 1)

The system according to claim 1, wherein the metadata includes a count of times that the historic data has been accessed by a user, an importance score associated with a location where the historic data was captured, an importance score associated with a device that captured the historic data, an importance score associated with a date-time stamp of the historic data, or a combination of the foregoing.

### Claim 6 (depends on 1)

The system according to claim 1, wherein the metadata includes information regarding visual features, aural features, or textual features of the historic data.

### Claim 7 (depends on 1)

The system according to claim 1, the system further comprising a searchable database and wherein the electronic computing device is further configured to
send the historic data with the updated metadata to the searchable database.

### Claim 8 (depends on 1)

The system according to claim 1, wherein the electronic computing device is further configured to
receive the high capacity machine learning model for detecting the object of interest.

### Claim 9 (depends on 1)

The system according to claim 1, wherein the electronic computing device is configured to
train the low capacity machine learning model to determine whether the object of interest is included in the historic data based on the metadata associated with the historic data.

### Claim 10 (depends on 1)

The system according to claim 1, wherein the electronic computing device is configured to
train the high capacity machine learning model to determine whether the object of interest is included in new data or the historic data based on characteristics of the new data or the historic data.

### Claim 11 (depends on 1)

The system according to claim 1, wherein a cost associated with training and running the low capacity machine learning model is lower than a cost associated with training and running the high capacity machine learning model.

### Claim 12 (depends on 1)

The system according to claim 1, wherein the high capacity machine learning model is more accurate than the low capacity machine learning model at detecting the object of interest in the historic data.

### Claim 13

A method for updating metadata associated with historic data, the method comprising
receiving, with an electronic computing device, historic data and metadata associated with the historic data;
using a low capacity machine learning model, analyzing the metadata associated with the historic data to determine a probability that the historic data includes an object of interest;
comparing the probability to a predetermined threshold; and
when the probability is greater than or equal to the predetermined threshold,
using a high capacity machine learning model to analyze the historic data to determine whether the historic data includes the object of interest; and
updating the metadata of the historic data based on whether the historic data includes the object of interest to generate updated metadata.

### Claim 14 (depends on 13)

The method according to claim 13, the method further comprising
when the probability is less than the predetermined threshold, using the low capacity machine learning model to update the metadata of the historic data based on whether the historic data includes the object of interest to generate updated metadata.

### Claim 15 (depends on 13)

The method according to claim 13, the method further comprising
receiving new data;
using the high capacity machine learning model, determining metadata associated with the new data, wherein the metadata associated with the new data includes whether or not the new data includes the object of interest; and
using the metadata associated with the new data and the metadata associated with the historic data to train the low capacity machine learning model.

### Claim 16 (depends on 13)

The method according to claim 13, wherein the historic data includes audio data, visual data, textual data, or a combination of the foregoing.

### Claim 17 (depends on 13)

The method according to claim 13, wherein the metadata includes a count of times that the historic data has been accessed by a user, an importance score associated with a location where the historic data was captured, an importance score associated with a device that captured the historic data, an importance score associated with a date-time stamp of the historic data, or a combination of the foregoing.

### Claim 18 (depends on 13)

The method according to claim 13, wherein the metadata includes information regarding visual features, aural features, or textual features of the historic data.

### Claim 19 (depends on 13)

The method according to claim 13, the method further comprising
sending the historic data with the updated metadata to a searchable database.

### Claim 20 (depends on 13)

The method according to claim 13, the method further comprising
receiving the high capacity machine learning model for detecting the object of interest.
