# US 2026/0099524 A1 · Computerized systems and methods for semantic searching

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

- Number: US 2026/0099524 A1
- Document ID: US20260099524A1
- Status: Published application
- Kind: A1
- Application number: 19/357,243
- Filed: 2025-10-14
- Published: 2026-04-09
- Earliest claimed: 2022-03-29
- Assignees: Workday, Inc.
- Applicants: Workday, Inc.
- Inventors: Joseph CAINEY; Jyoti MISHRA; Jakob NIELSEN; Bjorn ARNHOLTZ; Gary ENGELBERT
- Web page: https://patentgrep.com/patent/US20260099524A1

## CPC

- G06F 16/334: Query execution (filtering based on additional data G06F16/335)
- G06F 16/335: Filtering based on additional data, e.g. user or group profiles (filtering in web context G06F16/9535, G06F16/9536)
- G06F 16/3326: {using relevance feedback from the user, e.g. relevance feedback on documents, documents sets, document terms or passages}
- G06F 16/3338: {Query expansion}
- G06F 16/3344: {using natural language analysis}
- G06F 16/338: Presentation of query results
- G06F 16/36: Creation of semantic tools, e.g. ontology or thesauri
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Disclosed are systems and methods for a semantic search framework that operates to provide a robust search feature of database hosted data. The disclosed framework (or tool) improves and expands how searches can be configured and executed. In some embodiments, the disclosed framework is configured to receive a search request for a term or phrase, contextualize it to a generally understood theme that is not hindered by language barriers, and leverage it in a manner that is able to retrieve the most relevant and accurate results. The disclosed framework can filter search requests in a manner that both expands its breadth while honing in on what is actually being requested. The disclosed framework can be embodied as computerized systems and methods that can topically search for content based on a query string (e.g., term or phrase), and output a results set that embodies the theme of a survey's feedback.

## Claims

### Claim 1

A method comprising:
receiving, by a device, a search request comprising a character string, the character string having a context with a defined meaning;
converting, by the device, the search request into a word embedding, the word embedding comprising information that represents the defined meaning;
comparing, by the device, the word embedding against a plurality of word embeddings;
determining, by the device, a similarity measure between the word embedding and each of the plurality of word embeddings;
comparing, by the device, each determined similarity measure to a similarity threshold;
identifying, by the device, a set of word embeddings, wherein each word embedding in the set of word embeddings has a similarity measure at least satisfying the similarity threshold;
selecting, by the device, a subset of word embeddings from the set of word embeddings;
identifying, by the device, a set of terms that corresponds to the subset of word embeddings; and
outputting, by the device, for display within a user interface, the set of terms.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
analyzing, by the device, the search request; and
identifying, by the device, data and metadata related to the search request, wherein a conversion of the search request is based on the identified data and metadata.

### Claim 3 (depends on 2)

The method of claim 2, wherein the data and metadata related to the search request corresponds to at least one of the defined meaning, an identity (ID) of a user associated with the search request, a time stamp, location, position (or title) of the user, network address of a device of the user, access rights of the user and length of search request.

### Claim 4 (depends on 1)

The method of claim 1, wherein the plurality of word embeddings relate to at least one of feedback from a survey and comments provided by a respondent to a survey.

### Claim 5 (depends on 1)

The method of claim 1, further comprising using a fixed value as the similarity threshold.

### Claim 6 (depends on 1)

The method of claim 1, further comprising determining the similarity threshold by:
generating a set of sub-bands, each sub-band in the set of sub-bands associated with a minimum and maximum similarity and a subset of the plurality of word embeddings;
selecting random word embeddings from each sub-band in the set of sub-bands;
receiving a selection of a random word embedding from the random word embeddings; and
using a similarity measurement of the random word embedding to generate the similarity threshold.

### Claim 7 (depends on 1)

The method of claim 1, further comprising determining the similarity threshold by inputting the word embedding of a search phrase or word embedding of a document into a predictive model and using an output of the predictive model as the similarity threshold.

### Claim 8

A non-transitory computer-readable storage medium for tangibly storing computer program instructions capable of being executed by a computer processor, the computer program instructions defining steps of:
receiving a search request comprising a character string, the character string having a context with a defined meaning;
converting the search request into a word embedding, the word embedding comprising information that represents the defined meaning;
comparing the word embedding against a plurality of word embeddings;
determining a similarity measure between the word embedding and each of the plurality of word embeddings;
comparing each determined similarity measure to a similarity threshold;
identifying a set of word embeddings, wherein each word embedding in the set of word embeddings has a similarity measure at least satisfying the similarity threshold;
selecting a subset of word embeddings from the set of word embeddings;
identifying a set of terms that corresponds to the subset of word embeddings; and
outputting for display within a user interface, the set of terms.

### Claim 9 (depends on 8)

The non-transitory computer-readable storage medium of claim 8, the steps further comprising:
analyzing the search request; and
identifying data and metadata related to the search request, wherein a conversion of the search request is based on the identified data and metadata.

### Claim 10 (depends on 9)

The non-transitory computer-readable storage medium of claim 9, wherein the data and metadata related to the search request corresponds to at least one of the defined meaning, an identity (ID) of a user associated with the search request, a time stamp, location, position (or title) of the user, network address of a device of the user, access rights of the user and length of search request.

### Claim 11 (depends on 8)

The non-transitory computer-readable storage medium of claim 8, wherein the plurality of word embeddings relate to at least one of feedback from a survey and comments provided by a respondent to a survey.

### Claim 12 (depends on 8)

The non-transitory computer-readable storage medium of claim 8, the steps further comprising using a fixed value as the similarity threshold.

### Claim 13 (depends on 8)

The non-transitory computer-readable storage medium of claim 8, the steps further comprising determining the similarity threshold by:
generating a set of sub-bands, each sub-band in the set of sub-bands associated with a minimum and maximum similarity and a subset of the plurality of word embeddings;
selecting random word embeddings from each sub-band in the set of sub-bands;
receiving a selection of a random word embedding from the random word embeddings; and
using a similarity measurement of the random word embedding to generate the similarity threshold.

### Claim 14 (depends on 8)

The non-transitory computer-readable storage medium of claim 8, the steps further comprising determining the similarity threshold by inputting the word embedding into a predictive model and using an output of the predictive model as the similarity threshold.

### Claim 15

A device comprising:
a processor configured to:
receive a search request comprising a character string, the character string having a context with a defined meaning;
convert the search request into a word embedding, the word embedding comprising information that represents the defined meaning;
compare the word embedding against a plurality of word embeddings;
determine a similarity measure between the word embedding and each of the plurality of word embeddings;
compare each determined similarity measure to a similarity threshold;
identify a set of word embeddings, wherein each word embedding in the set of word embeddings has a similarity measure at least satisfying the similarity threshold;
select a subset of word embeddings from the set of word embeddings;
identify a set of terms that corresponds to the subset of word embeddings; and
output, for display within a user interface, the set of terms.

### Claim 16 (depends on 15)

The device of claim 15, the processor further configured to:
analyze the search request; and
identify data and metadata related to the search request, wherein a conversion of the search request is based on the identified data and metadata.

### Claim 17 (depends on 15)

The device of claim 15, wherein the plurality of word embeddings relate to at least one of feedback from a survey and comments provided by a respondent to a survey.

### Claim 18 (depends on 15)

The device of claim 15, the processor further configured to use a fixed value as the similarity threshold.

### Claim 19 (depends on 15)

The device of claim 15, the processor further configured to determine the similarity threshold by:
generating a set of sub-bands, each sub-band in the set of sub-bands associated with a minimum and maximum similarity and a subset of the plurality of word embeddings;
selecting random word embeddings from each sub-band in the set of sub-bands;
receiving a selection of a random word embedding from the random word embeddings; and
using a similarity measurement of the random word embedding to generate the similarity threshold.

### Claim 20 (depends on 15)

The device of claim 15, the processor further configured to determine the similarity threshold by inputting the word embedding into a predictive model and using an output of the predictive model as the similarity threshold.
