# US 2020/0401877 A1 · Maintaining master data using hierarchical classification

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

- Number: US 2020/0401877 A1
- Document ID: US20200401877A1
- Status: Published application
- Kind: A1
- Application number: 16/444,222
- Filed: 2019-06-18
- Published: 2020-12-24
- Earliest claimed: 2019-06-18
- Assignees: Sap SE
- Applicants: Sap SE
- Inventors: Francesco ALDA; Evgeny ARNAUTOV; Amrit RAJ; Sergey SMIRNOV; Ekaterina SUTTER
- Web page: https://patentgrep.com/patent/US20200401877A1

## CPC

- G06N 3/08: Learning methods
- G06F 16/258: {Data format conversion from or to a database}
- G06F 16/285: {Clustering or classification}
- G06N 3/0499: Feedforward networks
- G06N 3/09: Supervised learning
- G06N 3/0985: Hyperparameter optimisation; Meta-learning; Learning-to-learn
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Disclosed herein are system, method, and computer program product embodiments for classifying data objects using machine learning. In an embodiment, an artificial neural network may be trained to identify explained variable values corresponding to data object attributes. For example, the explained variables may be a category and a subcategory with the subcategory having a hierarchical relationship to the category. The artificial neural network may then receive a data record having one or more attribute values. The neural network may then identify a first and second explained variable value corresponding to the one or more attribute values based on the trained neural network model. The first and second explained variable values may then be associated with the data record. For example, if the data record is stored in a database, the record may be updated to include the first and second explained variable values.

## Claims

### Claim 1

A computer implemented method, comprising:
receiving a data record having one or more attribute values;
applying a neural network to the data record to identify a first value corresponding to a first explained variable and a second value corresponding to a second explained variable, the first value and the second value corresponding to the one or more attribute values and wherein the second explained variable has a hierarchical relationship to the first explained variable; and
associating the first value and the second value with the data record.

### Claim 2 (depends on 1)

The computer implemented method of claim 1, wherein the applying further comprises:
applying data preprocessing to the data record to standardize a format of the one or more attribute values.

### Claim 3 (depends on 1)

The computer implemented method of claim 1, wherein the applying further comprises:
embedding textual attribute values of the one or more attribute values to generate one or more vectors corresponding to the textual attribute values; and
concatenating the one or more vectors with numerical attribute values of the one or more attribute values.

### Claim 4 (depends on 3)

The computer implemented method of claim 3, wherein the applying further comprises:
embedding categorical attribute values of the one or more attribute values to generate one or more vectors corresponding to the categorical attribute values; and
concatenating the one or more vectors corresponding to the categorical attribute values with the numerical attribute values and the one or more vectors corresponding to the textual attribute values.

### Claim 5 (depends on 1)

The computer implemented method of claim 1, wherein the applying further comprises:
applying a first fully connected layer to identify the category; and
applying a second fully connected layer to identify the subcategory.

### Claim 6 (depends on 1)

The computer implemented method of claim 1, wherein the associating further comprises:
editing a database storing the data record to include the first value and the second value.

### Claim 7 (depends on 1)

The computer implemented method of claim 1, wherein the first explained variable is a category and the second explained variable is a subcategory of the category.

### Claim 8

A system, comprising:
a memory; and
at least one processor coupled to the memory and configured to:
receive a data record having one or more attribute values;
apply a neural network to the data record to identify a first value corresponding to a first explained variable and a second value corresponding to a second explained variable, the first value and the second value corresponding to the one or more attribute values and wherein the second explained variable has a hierarchical relationship to the first explained variable; and
associate the first value and the second value with the data record.

### Claim 9 (depends on 8)

The system of claim 8, wherein to apply the neural network, the at least one processor is further configured to:
apply data preprocessing to the data record to standardize a format of the one or more attribute values.

### Claim 10 (depends on 8)

The system of claim 8, wherein to apply the neural network, the at least one processor is further configured to:
embed textual attribute values of the one or more attribute values to generate one or more vectors corresponding to the textual attribute values; and
concatenate the one or more vectors with numerical attribute values of the one or more attribute values.

### Claim 11 (depends on 10)

The system of claim 10, wherein to apply the neural network, the at least one processor is further configured to:
embed categorical attribute values of the one or more attribute values to generate one or more vectors corresponding to the categorical attribute values; and
concatenate the one or more vectors corresponding to the categorical attribute values with the numerical attribute values and the one or more vectors corresponding to the textual attribute values.

### Claim 12 (depends on 8)

The system of claim 8, wherein to apply the neural network, the at least one processor is further configured to:
apply a first fully connected layer to identify the category; and
apply a second fully connected layer to identify the subcategory.

### Claim 13 (depends on 8)

The system of claim 8, wherein to associate the category and the subcategory to the data record, the at least one processor is further configured to:
edit a database storing the data record to include the first value and the second value.

### Claim 14 (depends on 8)

The system of claim 8, wherein the first explained variable is a category and the second explained variable is a subcategory of the category.

### Claim 15

A non-transitory computer-readable device having instructions stored thereon that, when executed by at least one computing device, cause the at least one computing device to perform operations comprising:
receiving a data record having one or more attribute values;
applying a neural network to the data record to identify a first value corresponding to a first explained variable and a second value corresponding to a second explained variable, the first value and the second value corresponding to the one or more attribute values and wherein the second explained variable has a hierarchical relationship to the first explained variable; and
associating the first value and the second value with the data record.

### Claim 16 (depends on 15)

The non-transitory computer-readable device of claim 15, wherein the applying comprises:
applying data preprocessing to the data record to standardize a format of the one or more attribute values.

### Claim 17 (depends on 15)

The non-transitory computer-readable device of claim 15, wherein the applying comprises:
embedding textual attribute values of the one or more attribute values to generate one or more vectors corresponding to the textual attribute values;
embedding categorical attribute values of the one or more attribute values to generate one or more vectors corresponding to the categorical attribute values; and
concatenating the one or more vectors corresponding to the categorical attribute values with the numerical attribute values and the one or more vectors corresponding to the textual attribute values.

### Claim 18 (depends on 15)

The non-transitory computer-readable device of claim 15, wherein the applying comprises:
applying a first fully connected layer to identify the category; and
applying a second fully connected layer to identify the subcategory.

### Claim 19 (depends on 15)

The non-transitory computer-readable device of claim 15, wherein the associating comprises:
editing a database storing the data record to include the first value and the second value.

### Claim 20 (depends on 15)

The non-transitory computer-readable device of claim 15, wherein the first explained variable is a category and the second explained variable is a subcategory of the category.