# US 12,387,478 B2 · Pattern recognition systems

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

- Number: US 12,387,478 B2
- Document ID: US12387478B2
- Status: Granted
- Kind: B2
- Application number: 18/613,859
- Filed: 2024-03-22
- Published: 2024-07-11
- Granted: 2025-08-12
- Earliest claimed: 2006-08-14
- Assignees: DataShapes, Inc.
- Applicants: DataShapes, Inc.
- Inventors: Jeffrey Brian Adams
- Web page: https://patentgrep.com/patent/US12387478B2

## CPC

- G06V 10/955: {using specific electronic processors}
- G06F 18/40: Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
- G06V 10/945: {User interactive design; Environments; Toolboxes}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Methods, apparatuses and systems directed to pattern identification and pattern recognition. In some particular implementations, the invention provides a flexible pattern recognition platform including pattern recognition engines that can be dynamically adjusted to implement specific pattern recognition configurations for individual pattern recognition applications. In some implementations, the present invention also provides for a partition configuration where knowledge elements can be grouped and pattern recognition operations can be individually configured and arranged to allow for multi-level pattern recognition schemes.

## Claims

### Claim 1

A system, comprising:
an interface configured to receive unstructured data;
a feature extraction module configured to extract a plurality of feature sets from the unstructured data;
a pattern recognition module configured to perform learning operations by which a knowledge base is generated based on first feature sets of the plurality of feature sets, the pattern recognition module also being configured to perform recognition operations in which second feature sets of the plurality of feature sets are compared to the knowledge base; and
memory configured to store the knowledge base;
wherein the interface, the feature extraction module, the pattern recognition module, and the memory are integrated in a single device.

### Claim 2 (depends on 1)

The system of claim 1, wherein each feature set represents a corresponding one of a plurality of portions of the unstructured data, each feature set also representing a point in a corresponding multi-dimensional space, and wherein the pattern recognition module is configured to perform the learning operations by generating a plurality of data objects from the feature sets, each data object corresponding to one or more of the feature sets, each data object representing a shape that includes the one or more points that the corresponding one or more feature sets represent.

### Claim 3 (depends on 2)

The system of claim 2, wherein the pattern recognition module is configured to perform the learning operations by associating metadata with each data object, the metadata for each data object identifying each portion of the unstructured data to which the data object corresponds.

### Claim 4 (depends on 3)

The system of claim 3, wherein the metadata for each data object also identifies a source of each portion of the unstructured data to which the data object corresponds.

### Claim 5 (depends on 1)

The system of claim 1, wherein the pattern recognition module is configured to perform the learning operations as supervised operations or unsupervised operations.

### Claim 6 (depends on 1)

The system of claim 1, wherein the pattern recognition module is configured to perform the recognition operations by comparing the second feature sets to knowledge elements of the knowledge base, wherein the knowledge elements of the knowledge base are organized in a plurality of partitions, each partition being configured to support a corresponding subset of the recognition operations.

### Claim 7 (depends on 6)

The system of claim 6, wherein the pattern recognition module is configured to perform the recognition operations by combining pattern recognition results from more than one of the partitions to generate a higher-level result.

### Claim 8 (depends on 1)

The system of claim 1, wherein the knowledge base includes a plurality of knowledge elements, and wherein the pattern recognition module is configured to maintain a count for each knowledge element, each count representing a number of the feature sets corresponding to that knowledge element, the counts collectively representing statistical regularities of patterns represented by the unstructured data.

### Claim 9 (depends on 1)

The system of claim 1, wherein the knowledge base includes a plurality of knowledge elements, and wherein the pattern recognition module is configured to perform the learning operations by associating metadata with each knowledge element, the metadata for each knowledge element identifying training data used in one or more of the learning operations associated with the knowledge element.

### Claim 10 (depends on 9)

The system of claim 9, wherein the pattern recognition module is configured to perform the recognition operations by retrieving the training data associated with a first knowledge element in response to a first recognition operation corresponding to the first knowledge element.

### Claim 11 (depends on 1)

The system of claim 1, wherein the knowledge base includes a plurality of knowledge elements, and wherein the pattern recognition module is configured to perform the learning operations by associating metadata with each knowledge element, the metadata for each knowledge element identifying remote data stored in a remote database separate from the knowledge base.

### Claim 12 (depends on 11)

The system of claim 11, wherein the pattern recognition module is configured to perform the recognition operations by retrieving the remote data associated with a first knowledge element in response to a first recognition operation corresponding to the first knowledge element.

### Claim 13 (depends on 1)

The system of claim 1, wherein different subsets of the feature sets include different numbers of features.

### Claim 14 (depends on 1)

The system of claim 1, wherein the unstructured data represent signals generated by one or more sensors, wherein the knowledge base includes a plurality of knowledge elements, and wherein the pattern recognition module is configured to perform the learning operations by associating metadata with each knowledge element, the metadata for each knowledge element identifying either or both of a corresponding sensor or corresponding sensor type.

### Claim 15 (depends on 1)

The system of claim 1, wherein the single device in which the interface, the feature extraction module, the pattern recognition module, and the memory are integrated is a controller for a data sensor.

### Claim 16

A method, comprising:
receiving unstructured data;
extracting a plurality of feature sets from the unstructured data;
performing learning operations by which a knowledge base is generated based on first feature sets of the plurality of feature sets;
performing recognition operations in which second feature sets of the plurality of feature sets are compared to the knowledge base; and
storing the knowledge base;
wherein receiving the unstructured data, extracting the feature sets, performing the learning operations, performing the recognition operations, and storing the knowledge base are all performed by a single device.

### Claim 17 (depends on 16)

The method of claim 16, wherein each feature set represents a corresponding one of a plurality of portions of the unstructured data, each feature set also representing a point in a corresponding multi-dimensional space, and wherein the performing the learning operations includes generating a plurality of data objects from the feature sets, each data object corresponding to one or more of the feature sets, each data object representing a shape that includes the one or more points that the corresponding one or more feature sets represent.

### Claim 18 (depends on 17)

The method of claim 17, wherein performing the learning operations includes associating metadata with each data object, the metadata for each data object identifying each portion of the unstructured data to which the data object corresponds.

### Claim 19 (depends on 18)

The method of claim 18, wherein the metadata for each data object also identifies a source of each portion of the unstructured data to which the data object corresponds.

### Claim 20 (depends on 16)

The method of claim 16, wherein the learning operations are supervised operations or unsupervised operations.

### Claim 21 (depends on 16)

The method of claim 16, wherein performing the recognition operations includes comparing the second feature sets to knowledge elements of the knowledge base, wherein the knowledge elements of the knowledge base are organized in a plurality of partitions, each partition being configured to support a corresponding subset of the recognition operations.

### Claim 22 (depends on 21)

The method of claim 21, wherein performing the recognition operations includes combining pattern recognition results from more than one of the partitions to generate a higher-level result.

### Claim 23 (depends on 16)

The method of claim 16, wherein the knowledge base includes a plurality of knowledge elements, and wherein the method further comprises maintaining a count for each knowledge element, each count representing a number of the feature sets corresponding to that knowledge element, the counts collectively representing statistical regularities of patterns represented by the unstructured data.

### Claim 24 (depends on 16)

The method of claim 16, wherein the knowledge base includes a plurality of knowledge elements, and wherein performing the learning operations includes associating metadata with each knowledge element, the metadata for each knowledge element identifying training data used in one or more of the learning operations associated with the knowledge element.

### Claim 25 (depends on 24)

The method of claim 24, wherein performing the recognition operations includes retrieving the training data associated with a first knowledge element in response to a first recognition operation corresponding to the first knowledge element.

### Claim 26 (depends on 16)

The method of claim 16, wherein the knowledge base includes a plurality of knowledge elements, and wherein performing the learning operations includes associating metadata with each knowledge element, the metadata for each knowledge element identifying remote data stored in a remote database separate from the knowledge base.

### Claim 27 (depends on 26)

The method of claim 26, wherein performing the recognition operations includes retrieving the remote data associated with a first knowledge element in response to a first recognition operation corresponding to the first knowledge element.

### Claim 28 (depends on 16)

The method of claim 16, wherein different subsets of the feature sets include different numbers of features.

### Claim 29 (depends on 16)

The method of claim 16, wherein the unstructured data represent signals generated by one or more sensors, wherein the knowledge base includes a plurality of knowledge elements, and wherein performing the learning operations includes associating metadata with each knowledge element, the metadata for each knowledge element identifying either or both of a corresponding sensor or corresponding sensor type.

### Claim 30 (depends on 16)

The method of claim 16, wherein the single device is a controller for a data sensor.