# US 2024/0202448 A1 · Automatic extraction of semantically similar question topics

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

- Number: US 2024/0202448 A1
- Document ID: US20240202448A1
- Status: Published application
- Kind: A1
- Application number: 18/065,758
- Filed: 2022-12-14
- Published: 2024-06-20
- Earliest claimed: 2022-12-14
- Assignees: Cisco Technology, Inc.
- Applicants: Cisco Technology, Inc.
- Inventors: Marek Suppa; Daniela Jass; Katarína Kozáková; Daniel Skala
- Web page: https://patentgrep.com/patent/US20240202448A1

## CPC

- G06F 40/30: Semantic analysis
- G06F 16/35: Clustering; Classification
- G06F 40/284: Lexical analysis, e.g. tokenisation or collocates
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A method, computer system, and computer program product are provided for automatically extracting semantically-similar question topics. A set of documents, wherein each document includes a plurality of words. One or more clusters of documents are identified in the set of documents based on a presence of common words in the documents of the one or more clusters. The one or more clusters are adjusted based on semantic similarity by adding or removing one or more documents from the one or more clusters. A topic is extracted from each adjusted cluster of documents.

## Claims

### Claim 1

A computer-implemented method comprising:
obtaining a set of documents, wherein each document includes a plurality of words;
identifying one or more clusters of documents in the set of documents based on a presence of common words in the documents of the one or more clusters;
adjusting the one or more clusters based on semantic similarity by adding or removing one or more documents from the one or more clusters; and
extracting a topic from each adjusted cluster of documents.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein each document is a question submitted by a participant of a question-and-answer session.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein extracting the topic from a particular cluster includes identifying a particular document having a longest sequence of stemmed words in common with a majority of other documents of the particular cluster.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, wherein adding or removing the one or more documents from the one or more clusters comprises generating a representation of the set of documents in a vector space, identifying a centroid for each cluster in the vector space, and adding or removing the one or more documents based on the one or more documents being within a threshold distance from the centroid.

### Claim 5 (depends on 4)

The computer-implemented method of claim 4, further comprising:
obtaining user feedback in response to presenting the extracted topic to one or more users; and
adjusting a value for the threshold distance based on the user feedback.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, further comprising presenting the extracted topic to one or more users via a display.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein the plurality of words in each document are tokenized prior to identifying the one or more clusters of documents.

### Claim 8

A computer system comprising:
one or more computer processors;
one or more computer readable storage media; and
program instructions stored on the one or more computer readable storage media for execution by at least one of the one or more computer processors, the program instructions comprising instructions to:
obtain a set of documents, wherein each document includes a plurality of words;
identify one or more clusters of documents in the set of documents based on a presence of common words in the documents of the one or more clusters;
adjust the one or more clusters based on semantic similarity by adding or removing one or more documents from the one or more clusters; and
extract a topic from each adjusted cluster of documents.

### Claim 9 (depends on 8)

The computer system of claim 8, wherein each document is a question submitted by a participant of a question-and-answer session.

### Claim 10 (depends on 8)

The computer system of claim 8, wherein the instructions to extract the topic from a particular cluster include instructions to identify a particular document having a longest sequence of stemmed words in common with a majority of other documents of the particular cluster.

### Claim 11 (depends on 8)

The computer system of claim 8, wherein adding or removing the one or more documents from the one or more clusters comprises generating a representation of the set of documents in a vector space, identifying a centroid for each cluster, and adding or removing the one or more documents based on the one or more documents being within a threshold distance from the centroid.

### Claim 12 (depends on 11)

The computer system of claim 11, wherein the instructions further comprise instructions to:
obtain user feedback in response to presenting the extracted topic to one or more users; and
adjust a value for the threshold distance based on the user feedback.

### Claim 13 (depends on 8)

The computer system of claim 8, further comprising instructions to present the extracted topic to one or more users via a display.

### Claim 14 (depends on 8)

The computer system of claim 8, wherein the plurality of words in each document are tokenized prior to identifying the one or more clusters of documents.

### Claim 15

A computer program product comprising one or more computer readable storage media collectively having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform operations including:
obtain a set of documents, wherein each document includes a plurality of words;
identify one or more clusters of documents in the set of documents based on a presence of common words in the documents of the one or more clusters;
adjust the one or more clusters based on semantic similarity by adding or removing one or more documents from the one or more clusters; and
extract a topic from each adjusted cluster of documents.

### Claim 16 (depends on 15)

The computer program product of claim 15, wherein each document is a question submitted by a participant of a question-and-answer session.

### Claim 17 (depends on 15)

The computer program product of claim 15, wherein the program instructions to extract the topic from a particular cluster further cause the computer to identify a particular document having a longest sequence of stemmed words in common with a majority of other documents of the particular cluster.

### Claim 18 (depends on 15)

The computer program product of claim 15, wherein adding or removing the one or more documents from the one or more clusters comprises generating a representation of the set of documents in a vector space, identifying a centroid for each cluster, and adding or removing the one or more documents based on the one or more documents being within a threshold distance from the centroid.

### Claim 19 (depends on 18)

The computer program product of claim 18, wherein the program instructions further cause the computer to:
obtain user feedback in response to presenting the extracted topic to one or more users; and
adjust a value for the threshold distance based on the user feedback.

### Claim 20 (depends on 15)

The computer program product of claim 15, wherein the program instructions further cause the computer to present the extracted topic to one or more users via a display.
