Chat-powered search using artificial intelligence

Patent metadata

Field	Value
Publication number	US-12524449-B2
Application number	US-202418672576-A
Country	US
Kind code	B2
Filing date	May 23, 2024
Priority date	May 23, 2024
Publication date	Jan 13, 2026
Grant date	Jan 13, 2026

Abstract

Methods, systems, and storage media for refining search queries through interactive conversational exchange are disclosed. Exemplary implementations may receive a first search input from a user describing a desired asset such as an image. Aspects implementations may also interact with the user via a chat interface to solicit additional details about the desired asset in response to the first search input, construct a search query based on the first search input and the additional details solicited from the user, and submit the search query to a search service to retrieve relevant asset data for the user.

First claim

What is claimed is:

1 . A method for refining search queries through an interactive conversational exchange, comprising:
receiving, from a user at a client device, a first search input describing a desired asset; interacting with the user via a chat interface in a conversational exchange to solicit additional details about the desired asset using prompts generated by a first large language model (LLM), the additional details comprising at least a first description of a desired setting of the desired asset; constructing, based on the conversational exchange including at least the first search input and the additional details, a complex-targeted search query to reflect complex user intents using a second LLM, wherein the complex user intents involve semantic meaning derived from user responses to the prompts provided to user via the chat interface; identifying, using a search service, a plurality of assets associated with the search query, wherein the search service is configured to handle queries with the complex user intents; and presenting the plurality of assets on a display screen of the client device, the display screen including at least the chat interface and results of the search service comprising the plurality of assets, the results being dynamically updated in response to user inputs received via the chat interface.

2 . The method of claim 1 , further comprising:
submitting the search query to the search service to retrieve asset data; generating an embedding based on the search query; and retrieving, using the embedding, the plurality of assets that are relevant to the search query from a database containing assets with associated embeddings.

3 . The method of claim 1 , further comprising prompting the user with specific questions related to characteristics of the desired asset, wherein the questions are formulated by a large language model based on the first search input and the additional details solicited from the user.

4 . The method of claim 1 , further comprising refining the search query in response to a second user input from the user describing additional details about the desired asset, wherein results from the refined search query are provided for presentation to the user.

5 . The method of claim 1 , wherein the user is enabled to provide feedback on the presented plurality of assets to further refine the search query.

6 . The method of claim 1 , further comprising constructing a chat history log that records the interactive conversational exchange, wherein the chat history log is utilized to inform the construction of the search query.

7 . The method of claim 1 , wherein the chat interface includes a visual representation of the interactive conversational exchange, and wherein the visual representation is updated in real-time as the interactive conversational exchange progresses.

8 . The method of claim 1 , wherein the additional details solicited from the user include a second description of a composition within the desired asset and a third description of a mood of the desired asset, wherein the first and second descriptions are incorporated into the search query.

9 . The method of claim 1 , wherein the user is prompted to provide feedback on specific attributes of the desired asset, and wherein the specific attributes include one or more of color scheme, lighting conditions, or presence of specific objects or characters.

10 . A system configured for refining search queries through an interactive conversational exchange, the system comprising:
one or more hardware processors configured by machine-readable instructions to:
receive, from a user at a client device, a first search input describing a desired asset;
interact with the user via a chat interface in a conversational exchange to solicit additional details about the desired asset, the additional details comprising at least a first description of a desired setting of the desired asset;
construct, based on the conversational exchange including at least the first search input and the additional details, a complex-targeted search query to reflect complex user intents, wherein the complex user intents involve semantic meaning derived from user responses to prompts provided to user via the chat;
identify, using a search service, a plurality of assets associated with the search query, wherein the search service is configured to handle queries with the complex user intents; and
present the plurality of assets on a display screen of the client device, the display screen including at least the chat interface and results of the search service comprising the plurality of assets, the results being dynamically updated in response to user inputs received via the chat interface.

11 . The system of claim 10 , wherein the one or more hardware processors are further configured by machine-readable instructions to:
submit the search query to the search service to retrieve asset data; generate an embedding based on the search query; and retrieve, using the embedding, the plurality of assets that are relevant to the search query from a database containing assets with associated embeddings.

12 . The system of claim 10 , wherein the one or more hardware processors are further configured by machine-readable instructions to prompt the user with specific questions related to characteristics of the desired asset, wherein the questions are formulated by a large language model based on the first search input and the additional details solicited from the user.

13 . The system of claim 10 , wherein the one or more hardware processors are further configured by machine-readable instructions to refine the search query in response to a second user input from the user describing additional details about the desired asset, wherein results from the refined search query are provided for presentation to the user.

14 . The system of claim 10 , wherein the user is enabled to provide feedback on the presented plurality of assets to further refine the search query.

15 . The system of claim 10 , wherein the one or more hardware processors are further configured by machine-readable instructions to construct a chat history log that records the interactive conversational exchange, wherein the chat history log is utilized to inform the construction of the search query.

16 . The system of claim 10 , wherein the chat interface includes a visual representation of the interactive conversational exchange, and wherein the visual representation is updated in real-time as the interactive conversational exchange progresses.

17 . The system of claim 10 , wherein the user is prompted to provide feedback on specific attributes of the desired asset, and wherein the specific attributes include one or more of color scheme, lighting conditions, or presence of specific objects or characters.

18 . A non-transient computer-readable storage medium having instructions embodied thereon, the instructions being executable by one or more processors to perform a method for refining search queries through an interactive conversational exchange, the method comprising:
receiving, from a user at a client device, a first search input describing a desired asset; interacting with the user via a chat interface in a conversational exchange to solicit additional details about the desired asset, the additional details comprising at least a first description of a desired setting of the desired asset; constructing, based on the conversational exchange including at least the first search input and the additional detai

Assignees

• Shutterstock Inc

Inventors

• Smyth Raymond Lúí
• Loy Christopher
• Kadam Keenan
• Williams Timothy

Classifications

• G06F16/3344

  using natural language analysis · CPC title

• G06F16/9535

  Search customisation based on user profiles and personalisation · CPC title

• G06F16/3328

  using graphical result space presentation or visualisation · CPC title

• G06F16/3329 Primary

  Natural language query formulation · CPC title

• G06F16/33295 Primary

  in dialogue systems · CPC title