# US 2026/0093742 A1 · Content-based annotation detection and summarization for documents

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

- Number: US 2026/0093742 A1
- Document ID: US20260093742A1
- Status: Published application
- Kind: A1
- Application number: 18/903,767
- Filed: 2024-10-01
- Published: 2026-04-02
- Earliest claimed: 2024-10-01
- Assignees: Open Text Holdings, Inc.
- Applicants: Open Text Holdings, Inc.
- Inventors: Samiksha Kharbanda; Ranjit Boddu
- Web page: https://patentgrep.com/patent/US20260093742A1

## CPC

- G06F 16/3347: {using vector based model}
- G06F 16/345: {Summarisation for human users}
- G06F 16/93: Document management systems
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Systems and methods for context-based annotation analyzation are provided. A system and method provides an artificial intelligence (AI) solution that aims to automate, accelerate, and enhance the accuracy of managing annotations and comments, thereby minimizing the risks associated with manual processing. The system and method leverages AI models to contextually identify and understand annotations within documents, allowing for precise summarization and enabling advanced search capabilities based on these annotations.

## Claims

### Claim 1

A method of context-based annotation analyzation, the method comprising:
obtaining a plurality of documents having annotations;
for each of the plurality of documents:
extracting annotation data and corresponding coordinate data, the coordinate data associating respective annotation data and text with at least one corresponding coordinate in a respective document,
creating a page chunk for the respective document, the page chunk combining the annotation data and the text for the respective document, and
creating an embedding from the annotation data and the text of the respective page chunk and storing the embedding in a vector store;
in response to a user query, providing relevant embeddings to a Large Language Model (LLM) trained to recognize patterns in annotations; and
generating a response to the user query using an output of the LLM.

### Claim 2 (depends on 1)

The method of claim 1, wherein a first document of the plurality of documents is an image document, and wherein the coordinate data identifies a location on the first document.

### Claim 3 (depends on 2)

The method of claim 2, wherein the coordinate data comprises x, y coordinates of the image document.

### Claim 4 (depends on 1)

The method of claim 1, wherein a first document of the plurality of documents is a text document, and wherein the coordinate data identifies a location in the text document.

### Claim 5 (depends on 4)

The method of claim 4, wherein a first document of the plurality of documents is a text document, and wherein the coordinate data identifies character positions in the text document.

### Claim 6 (depends on 1)

The method of claim 1, wherein the relevant embeddings are determined by a similarity search.

### Claim 7 (depends on 6)

The method of claim 6, wherein the similarity search determines similarities between the user query and the embeddings stored in the vector store.

### Claim 8

A system for context-based annotation analyzation, the system comprising:
a processor; and
a non-transitory computer readable medium storing instructions translatable by the processor, the instructions when translated by the processor perform:
obtaining a plurality of documents having annotations;
for each of the plurality of documents:
extracting annotation data and corresponding coordinate data, the coordinate data associating respective annotation data and text with at least one corresponding coordinate in a respective document,
creating a page chunk for the respective document, the page chunk combining the annotation data and the text for the respective document, and
creating an embedding from the annotation data and the text of the respective page chunk and storing the embedding in a vector store;
in response to a user query, providing relevant embeddings to a Large Language Model (LLM) trained to recognize patterns in annotations; and
generating a response to the user query using an output of the LLM.

### Claim 9 (depends on 8)

The system of claim 8, wherein a first document of the plurality of documents is an image document, and wherein the coordinate data identifies a location on the first document.

### Claim 10 (depends on 9)

The system of claim 9, wherein the coordinate data comprises x, y coordinates of the image document.

### Claim 11 (depends on 8)

The system of claim 8, wherein a first document of the plurality of documents is a text document, and wherein the coordinate data identifies a location in the text document.

### Claim 12 (depends on 11)

The system of claim 11, wherein a first document of the plurality of documents is a text document, and wherein the coordinate data identifies character positions in the text document.

### Claim 13 (depends on 8)

The system of claim 8, wherein the relevant embeddings are determined by a similarity search.

### Claim 14 (depends on 13)

The system of claim 13, wherein the similarity search determines similarities between the user query and the embeddings stored in the vector store.

### Claim 15

A computer program product comprising a non-transitory computer readable medium storing instructions translatable by a processor, the instructions when translated by the processor perform, in an enterprise computing network environment:
obtaining a plurality of documents having annotations;
for each of the plurality of documents:
extracting annotation data and corresponding coordinate data, the coordinate data associating respective annotation data and text with at least one corresponding coordinate in a respective document,
creating a page chunk for the respective document, the page chunk combining the annotation data and the text for the respective document, and
creating an embedding from the annotation data and the text of the respective page chunk and storing the embedding in a vector store;
in response to a user query, providing relevant embeddings to a Large Language Model (LLM) trained to recognize patterns in annotations; and
generating a response to the user query using an output of the LLM.

### Claim 16 (depends on 15)

The computer program product of claim 15, wherein a first document of the plurality of documents is an image document, and wherein the coordinate data identifies a location on the first document.

### Claim 17 (depends on 16)

The computer program product of claim 16, wherein the coordinate data comprises x, y coordinates of the image document.

### Claim 18 (depends on 15)

The computer program product of claim 15, wherein a first document of the plurality of documents is a text document, and wherein the coordinate data identifies a location in the text document.

### Claim 19 (depends on 15)

The computer program product of claim 15, wherein the relevant embeddings are determined by a similarity search.

### Claim 20 (depends on 19)

The computer program product of claim 19, wherein the similarity search determines similarities between the user query and the embeddings stored in the vector store.
