# US 2018/0322340 A1 · Continuously self-adapting arrangement for automatically acquiring data and corresponding method

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

- Number: US 2018/0322340 A1
- Document ID: US20180322340A1
- Status: Published application
- Kind: A1
- Application number: 15/962,887
- Filed: 2018-04-25
- Published: 2018-11-08
- Earliest claimed: 2017-05-05
- Assignees: Open Text SA ULC
- Applicants: Open Text SA Ulc
- Inventors: Rolf Jahn
- Web page: https://patentgrep.com/patent/US20180322340A1

## CPC

- G06N 20/00: Machine learning
- G06F 16/2455: Query execution
- G06F 16/355: Creation or modification of classes or clusters
- G06F 16/84: Mapping; Conversion
- G06F 16/93: Document management systems
- G06F 18/2178: {based on feedback of a supervisor}
- G06V 30/414: Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

The invention relates to a method for continuously self-adapting an arrangement for automatically acquiring data from a document using a model. The method comprises: receiving raw data of a document containing a content item, the raw data including document content information representing the content item; interpreting the received raw data based on the model to generate interpretation data including interpreted content information of the content item of the document; validating and correcting the interpretation data of the document using the raw data to generate correction data of the document, the correction data including corrected content information of the content item of the document; analyzing the interpretation data, the raw data and the correction data of the document to determine whether to update the model or not; and updating the model content information of the model using the corrected content information of the correction data.

## Claims

### Claim 1

A computer-implemented method, comprising:
receiving, by a computer, raw data of a document, the document containing a content item, the raw data including document content information representing the content item;
interpreting, by the computer using a model, the received raw data to generate interpretation data including interpreted content information of the content item of the document, the model including model content information representing the content item in a pattern document;
validating and correcting, by the computer using the raw data, the interpretation data of the document to generate correction data of the document, the correction data including corrected content information of the content item of the document;
analyzing, by the computer, the interpretation data, the raw data, and the correction data of the document to determine whether to update the model or not; and
updating, by the computer, the model content information of the model using the corrected content information of the correction data to thereby allow the computer to automatically self-adapt and improve automatic data acquisition.

### Claim 2 (depends on 1)

The computer-implemented method according to claim 1, further comprising:
processing the document by Optical Character Recognition to generate the raw data.

### Claim 3 (depends on 1)

The computer-implemented method according to claim 1, further comprising:
storing the interpretation data in association with the raw data of the document; and
searching for stored interpretation data and raw data corresponding to the correction data.

### Claim 4 (depends on 1)

The computer-implemented method according to claim 1, wherein the model further includes model zone information indicating a zone in the pattern document in which the content item of the pattern document is placed, and wherein the interpreting includes searching for, based on document zone information indicating a zone of the content item in the document, model zone information in the model indicating a zone which substantially coincides with the zone of the content item in the document.

### Claim 5 (depends on 1)

The computer-implemented method according to claim 1, wherein the validating comprises applying a predetermined analysis function corresponding to the model used for interpretation, interacting with an operator, or both.

### Claim 6 (depends on 1)

The computer-implemented method according to claim 1, further comprising:
identifying, using the raw data received, the pattern document among a plurality of pattern documents having a layout corresponding to a layout of the document; and
determining the model for the pattern document among a plurality of models.

### Claim 7 (depends on 1)

The computer-implemented method according to claim 1, wherein the model further includes at least one of model description information describing a pattern document, a model identifier uniquely identifying the model, layout information defining a layout of the pattern document, a keyword corresponding to the content item, or model position information indicating a position of the content item in the pattern document, wherein the model content information contains at least one of language information, format information, or description information.

### Claim 8

A system, comprising:
a processor;
a non-transitory computer-readable medium; and
stored instructions translatable by the processor for:
receiving raw data of a document, the document containing a content item, the raw data including document content information representing the content item;
interpreting, using a model, the received raw data to generate interpretation data including interpreted content information of the content item of the document, the model including model content information representing the content item in a pattern document;
validating and correcting, using the raw data, the interpretation data of the document to generate correction data of the document, the correction data including corrected content information of the content item of the document;
analyzing the interpretation data, the raw data, and the correction data of the document to determine whether to update the model or not; and
updating the model content information of the model using the corrected content information of the correction data to thereby allow the system to automatically self-adapt and improve automatic data acquisition.

### Claim 9 (depends on 8)

The system of claim 8, wherein the stored instructions are further translatable by the processor for:
processing the document by Optical Character Recognition to generate the raw data.

### Claim 10 (depends on 8)

The system of claim 8, wherein the stored instructions are further translatable by the processor for:
storing the interpretation data in association with the raw data of the document; and
searching for stored interpretation data and raw data corresponding to the correction data.

### Claim 11 (depends on 8)

The system of claim 8, wherein the model further includes model zone information indicating a zone in the pattern document in which the content item of the pattern document is placed, and wherein the interpreting includes searching for, based on document zone information indicating a zone of the content item in the document, model zone information in the model indicating a zone which substantially coincides with the zone of the content item in the document.

### Claim 12 (depends on 8)

The system of claim 8, wherein the validating comprises applying a predetermined analysis function corresponding to the model used for interpretation, interacting with an operator, or both.

### Claim 13 (depends on 8)

The system of claim 8, wherein the stored instructions are further translatable by the processor for:
identifying, using the raw data received, the pattern document among a plurality of pattern documents having a layout corresponding to a layout of the document; and
determining the model for the pattern document among a plurality of models.

### Claim 14 (depends on 8)

The system of claim 8, wherein the model further includes at least one of model description information describing a pattern document, a model identifier uniquely identifying the model, layout information defining a layout of the pattern document, a keyword corresponding to the content item, or model position information indicating a position of the content item in the pattern document, wherein the model content information contains at least one of language information, format information, or description information.

### Claim 15

A computer program product comprising a non-transitory computer-readable medium storing instructions translatable by a processor for:
receiving raw data of a document, the document containing a content item, the raw data including document content information representing the content item;
interpreting, using a model, the received raw data to generate interpretation data including interpreted content information of the content item of the document, the model including model content information representing the content item in a pattern document;
validating and correcting, using the raw data, the interpretation data of the document to generate correction data of the document, the correction data including corrected content information of the content item of the document;
analyzing the interpretation data, the raw data, and the correction data of the document to determine whether to update the model or not; and
updating the model content information of the model using the corrected content information of the correction data to thereby allow the computer program product to automatically self-adapt and improve automatic data acquisition.

### Claim 16 (depends on 15)

The computer program product of claim 15, wherein the instructions are further translatable by the processor for:
processing the document by Optical Character Recognition to generate the raw data.

### Claim 17 (depends on 15)

The computer program product of claim 15, wherein the instructions are further translatable by the processor for:
storing the interpretation data in association with the raw data of the document; and
searching for stored interpretation data and raw data corresponding to the correction data.

### Claim 18 (depends on 15)

The computer program product of claim 15, wherein the model further includes model zone information indicating a zone in the pattern document in which the content item of the pattern document is placed, and wherein the interpreting includes searching for, based on document zone information indicating a zone of the content item in the document, model zone information in the model indicating a zone which substantially coincides with the zone of the content item in the document.

### Claim 19 (depends on 15)

The computer program product of claim 15, wherein the validating comprises applying a predetermined analysis function corresponding to the model used for interpretation, interacting with an operator, or both.

### Claim 20 (depends on 15)

The computer program product of claim 15, wherein the model further includes at least one of model description information describing a pattern document, a model identifier uniquely identifying the model, layout information defining a layout of the pattern document, a keyword corresponding to the content item, or model position information indicating a position of the content item in the pattern document, wherein the model content information contains at least one of language information, format information, or description information.
