Performing targeted searching based on a user profile

Patent metadata

Field	Value
Publication number	US-12468715-B2
Application number	US-202418438863-A
Country	US
Kind code	B2
Filing date	Feb 12, 2024
Priority date	Jan 29, 2021
Publication date	Nov 11, 2025
Grant date	Nov 11, 2025

Abstract

Aspects of the present disclosure relate to systems and methods for performing targeted searching based on a user profile. In examples, a user profile including a user embedding may be retrieved based on the receipt of a user indication. The user embedding may be created based on one or more user interest. A plurality of document embeddings may be identified based on the user embedding, where each document embedding of the plurality of document embeddings is determined to be within a first distance of the user embedding. In examples, a ranking for each document embedding of the plurality of document embeddings may be generated, where the ranking for each document embedding of the plurality of document embeddings is based on the user embedding. At least one document may be recommend based on a ranking associated with a document embedding.

First claim

What is claimed is:

1 . A method for performing a targeted search based on user interests, the method comprising:
generating a user embedding as a low-dimensional space representation of a user based on user information via a first machine learning model; generating a plurality of document embeddings as low-dimensional space representations of documents via a second machine learning model; receiving a user identifier for the user; retrieving the user embedding based on the user identifier; ranking one or more document embeddings of the plurality of document embeddings based on the user embedding; and recommending to the user at least one document based on the ranking of the document embeddings.

2 . The method of claim 1 , wherein the user embedding further comprises a low-dimensional space representation of a translated higher-dimensional vector derived from the user information associated with the user.

3 . The method of claim 1 , wherein ranking one or more document embeddings of the plurality of document embeddings based on the user embedding comprises:
identifying a set of document embeddings from the plurality of document embeddings using the user embedding, each document embedding of the set of document embeddings determined to be within a first distance of the user embedding; and ranking each document embedding of the set of document embeddings based on the user embedding.

4 . The method of claim 3 , wherein ranking each document embedding of the set of document embeddings based on the user embedding comprises ranking each document embedding of the set of document embeddings using a third machine learning model based on one or more of relevancy, novelty, and diversity.

5 . The method of claim 1 , wherein the user information comprises the user identifier, user characteristics, user interests, and user actions, the user interests further comprise an indication of one or more topics of interest to the user and one or more topics that are not of interest to the user, and the user actions further comprise one or more previous documents viewed by the user.

6 . The method of claim 1 , further comprising:
extracting location and content information from a document; providing the location and content information for the document to a fourth machine learning model; generating, via the fourth machine learning model, a document embedding specific to the document; and adding the document embedding to the plurality of document embeddings.

7 . The method of claim 1 , further comprising displaying the recommended at least one document at an output device associated with the user.

8 . The method of claim 7 , further comprising:
receiving an indication that another user selected the recommended at least one document displayed to the output device; generating another user embedding based on the received indication; identifying a second set of document embeddings from the plurality of document embeddings based on the another user embedding; generating a ranking for each document embedding of the second set of document embeddings, wherein the ranking for each document embedding of the second set of document embeddings is based on the another user embedding; and recommending to the user at least one document based on the ranking for each document embedding of the second set of document embeddings.

9 . A system for performing a targeted search based on user interests, the system comprising:
a processor; memory including instructions, which when executed by the processor, cause the processor to:
generate a user embedding as a low-dimensional space representation of a user based on user information via a first machine learning model;
generate a plurality of document embeddings as low-dimensional space representations of documents via a second machine learning model;
receive a user identifier for the user;
retrieve the user embedding based on the user identifier;
rank one or more document embeddings of the plurality of document embeddings based on the user embedding; and
recommend to the user at least one document based on the ranking of the document embeddings.

10 . The system of claim 9 , wherein to rank one or more document embeddings of the plurality of document embeddings based on the user embedding comprises to:
identify a set of document embeddings from the plurality of document embeddings using the user embedding, each document embedding of the set of document embeddings determined to be within a first distance of the user embedding; and rank each document embedding of the set of document embeddings based on the user embedding.

11 . The system of claim 10 , wherein to rank each document embedding of the set of document embeddings based on the user embedding comprises to rank each document embedding of the set of document embeddings using a third machine learning model based on one or more of relevancy, novelty, and diversity.

12 . The system of claim 10 , wherein the user information comprises the user identifier, user characteristics, user interests, and user actions, the user interests further comprise an indication of one or more topics of interest to the user and one or more topics that are not of interest to the user, and the user actions further comprise one or more previous documents viewed by the user.

13 . The system of claim 9 , wherein the instructions, which when executed by the processor, cause the processor to:
extract location and content information from a target document; provide the location and content information for the target document to a fourth machine learning model; generate, via the fourth machine learning model, a target document embedding specific to the target document; and add the target document embedding to the plurality of document embeddings.

14 . The system of claim 9 , wherein the instructions, which when executed by the processor, cause the processor to display the recommended at least one document at an output device associated with the user.

15 . A computer storage medium including instructions, which when executed by a processor, cause the processor to:
generate a user embedding as a low-dimensional space representation of a user based on user information via a first machine learning model; generate a plurality of document embeddings as low-dimensional space representations of documents via a second machine learning model; receive a user identifier for the user; retrieve the user embedding based on the user identifier; rank one or more document embeddings of the plurality of document embeddings based on the user embedding; and recommend to the user at least one document based on the ranking of the document embeddings.

16 . The computer storage medium of claim 15 , wherein to rank one or more document embeddings of the plurality of document embeddings based on the user embedding comprises to:
identify a set of document embeddings from the plurality of document embeddings using the user embedding, each document embedding of the set of document embeddings determined to be within a first distance of the user embedding; and rank each document embedding of the set of document embeddings based on the user embedding.

17 . The computer storage medium of claim 15 , wherein the user information comprises the user identifier, user characteristics, user interests, and user actions, the user interests further comprise an indication of one or more topics of interest to the user and one or more topics that are not of interest to the user, and the user actions further compr

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Ahmed Junaid
• Malik Waleed
• Overwijk Arnold

Classifications

• G06N20/00

  Machine learning · CPC title

• G06F16/24578 Primary

  using ranking · CPC title

• G06F16/9035 Primary

  Filtering based on additional data, e.g. user or group profiles · CPC title