Identifying products completing tasks of search query definitions using generative artificial intelligence

Abstract

The invention relates to a product for completing a task of search query definition using generative artificial intelligence identification. A search engine utilizes Generative Artificial Intelligence (AI) to identify products in order to complete tasks defined by a search query. A user having a question or task in hand initially inputs a search query including an intent or environment describing the question or task. The search engine determines that the search query is an intent-based query, rather than performing a search based on the search query. The search engine may further determine that the search query is not a cached query. The search query is provided to a generative AI model, and a list of items corresponding to tasks defined by the search query is generated. The list of items may be provided to the user as search suggestions. The user may select a portion of the item from the list of items and perform a search.

Description

Technical Field

[0001] The present invention relates to utilizing generative artificial intelligence to identify products that accomplish tasks defined by search queries. Background Art

[0002] Many product search systems allow users to submit search queries consisting of several words or terms. The search system returns a list of relevant items associated with the search query based on keyword searches available within the corresponding site. If the search terms do not match the item title or item description, the search system may not return appropriate results. Summary of the invention

[0003] At a high level, various aspects of the present disclosure relate to intent-based search. More specifically, various aspects of the present disclosure relate to a search engine that utilizes generative artificial intelligence (AI) to identify products for completing a task defined by a search query. According to various aspects of the technology described in the present disclosure, a search query is received at a search engine. Based on determining that the search query is an intent-based query, the search query is provided to a generative AI model. A list of items corresponding to the task defined by the search query is received at the search engine. The search engine performs a search on at least a portion of the list of items and provides search results to a user.

[0004] This summary is intended to introduce in simplified form some concepts that are further described in the embodiments of the present disclosure. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used to help determine the scope of the claimed subject matter. Other objects, advantages, and novel features of the present technology will be provided, and some of these objects, advantages, and novel features will become apparent to those skilled in the art upon examination of the present disclosure or learning through practicing the present technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] The patent or application file contains at least one drawing in color. Copies of this patent or patent application publication with color drawings will be provided by the Patent Office upon request and payment of the necessary fee. The technology is described in detail below with reference to the accompanying drawings, in which:

[0006] Figure 1 is a block diagram of an example operating environment suitable for implementing aspects of the present technology;

[0007] Figure 2 is an example user interface showing search results before a search system utilizes generative AI according to one aspect of the technology described herein;

[0008] Figure 3 is an example user interface showing search results after a search system utilizes generative AI according to one aspect of the technology described herein;

[0009] Figure 4 is an example user interface showing search results before a search system utilizes generative AI according to one aspect of the technology described herein;

[0010] Figure 5 is an example user interface showing search results after a search system utilizes generative AI according to one aspect of the technology described herein;

[0011] Figure 6 is a flow chart illustrating a method for utilizing generative AI to identify products for completing a task according to an aspect of the technology described herein;

[0012] Figure 7 is a flow chart illustrating a method for utilizing generative AI to identify products for completing a task according to an aspect of the technology described herein; and

[0013] Figure 8 is an example computing device suitable for implementing the described techniques according to one aspect described herein. DETAILED DESCRIPTION

[0014] The subject matter of various aspects of the present technology is specifically described herein to meet statutory requirements. However, the description itself is not intended to limit the scope of the present disclosure. On the contrary, the inventors have contemplated that the claimed subject matter may also be implemented in other ways, thereby in combination with other existing or future technologies to include different steps or combinations of steps similar to those described herein. In addition, although the terms "step" and / or "block" may be used herein to refer to different elements of the method employed, these terms should not be interpreted as implying any particular order between the various steps disclosed herein unless the order of the various steps is explicitly described.

[0015] Furthermore, words such as "a" and "an" may also include the plural as well as the singular unless otherwise indicated. Thus, for example, where there are one or more features, the limitation of "feature" is satisfied. Furthermore, the term "or" includes conjunction, disjunction, and both (thus a or b includes a or b, and a and b).

[0016] Although search engines are very useful tools for providing search results for received search queries, shortcomings in existing search technologies often result in the consumption of unnecessary amounts of computing resources (e.g., I / O costs, network packet generation costs, throughput, memory consumption, etc.). When performing a search, users often look for specific search results. For example, in a product-based search environment, a user may be looking for a specific item with a specific style, design, or color. However, in an intent-based search environment, a user may be looking for multiple items (some of which may not be known to the user) to perform a specific task. In this environment, existing search technologies cannot understand the user's intent, so the search results are not helpful. This requires the user to submit additional search queries or multiple filters to obtain the desired search results related to the items required to perform the task.

[0017] For example, a search engine may receive a text-based query "repair torn clothes". Because the search system cannot understand the user's intent, the search system returns results that include items available within the e-commerce website (e.g., clothes). However, in this example, the user is searching for items that are used to actually repair torn clothes (e.g., ironing patches, fabric glue, sewing kits, fabric tape, etc.), and the results provided are not helpful. In another example, a search engine may receive a text-based query "necessities for travel". Because the search system cannot understand the user's intent, the search system returns results for items available within the site that have titles or descriptions that match the search terms (e.g., books with words such as "necessities", "for", and "travel" in the title or description). However, in this example, the user is searching for items needed for travel (e.g., backpacks, sleeping bags, sunscreen, cameras, etc.), and, again, the results provided are not helpful.

[0018] This requires the user to perform multiple searches to find out the items that are actually needed to perform the task at hand, and to perform multiple searches to find items that are available for purchase. This process unnecessarily consumes various computing resources of the search system, such as processing power, network bandwidth, throughput, memory consumption, etc. In some instances, multiple attempts to identify the items required to perform the task may not even satisfy the user's goal at all, thus requiring the user to spend more time and computing resources on the search process by repeating the process of issuing additional queries until the user finally accesses the desired satisfactory items. In some cases, the user may even abandon the search because the search engine fails to return the desired search results after multiple searches.

[0019] These shortcomings in existing search technologies have an adverse impact on computer network communications. For example, each time a query is received, the content or payload of the search query is typically supplemented with header information or other metadata that is multiplied by all the additional queries required to obtain the specific items that the user desires to complete the task. Therefore, by repeatedly generating this metadata and sending it over the computer network, throughput and latency costs are incurred. In some instances, these repeated inputs (e.g., repeated clicks, selections, or queries) increase the I / O of the storage device (e.g., excessive physical read / write head movement on a non-volatile disk) because each time the user enters unnecessary information, such as entering several queries, the computing system often has to contact the storage device to perform a read or write operation. This is time-consuming, error-prone, and will eventually wear out components such as read / write heads. In addition, if multiple users repeatedly issue queries, it will cost a lot because processing queries consumes a lot of computing resources. For example, for some search engines, it may be necessary to calculate a query execution plan each time a query is issued. This requires the search system to find the query execution plan with the least cost to fully execute the query. This reduces throughput, increases network latency, and can waste valuable time.

[0020] In view of these shortcomings in existing search technologies, aspects of the technology described herein improve the functionality of computers themselves by providing a solution that enables search engines to provide enhanced search accuracy by leveraging generative AI, enabling users to purchase items to complete tasks through a single search query. A machine learning model is utilized to generate a list of items from a text query. It can be appreciated that better results are obtained through the technology described herein compared to traditional search engines that require multiple search queries from users.

[0021] Various aspects of the technology described herein provide many improvements over existing search technologies. For example, computing resource consumption is improved relative to the prior art. Specifically, for intent-based queries, search accuracy is enhanced by leveraging generative AI, allowing users to obtain relevant search results more quickly. This eliminates (or at least reduces) repeated user queries and filter selections because the search results include a list of items that complete the task that the user is looking for. Therefore, various aspects of the technology described herein reduce computing resource consumption, such as processing power and network bandwidth. For example, a user query (e.g., an HTTP request) will only need to traverse a computer network once (or a fewer number of times relative to the prior art).

[0022] In a similar manner, various aspects of the technology described herein improve storage device or disk I / O and query execution functions. As described above, the insufficient search results provided by existing search technologies lead to repeated user queries and filter selections. This can result in multiple traversals of disk I / O. In contrast, various aspects described herein reduce storage device I / O because the input provided by the user is reduced, so the computing system does not have to frequently contact the storage device to perform read or write operations. For example, by utilizing generative AI, the search engine can respond with enhanced search results, enabling the user to purchase items through a single search query to complete the task. Therefore, there will not be too much wear and tear on the components due to the query execution function.

[0023] Having briefly described an overview of various aspects of the technology described herein, an exemplary operating environment in which various aspects of the technology described herein may be implemented is described below.

[0024] Now go to Figure 1 , Figure 1 A block diagram is provided showing an operating environment 100 in which various aspects of the present disclosure may be employed. It should be understood that this and other settings described herein are presented as examples only. In addition to or in place of the settings and elements shown, other settings and elements (e.g., machines, interfaces, functions, sequences, and functional groupings) may be used, and for clarity, some elements may be omitted completely. In addition, many elements described herein are functional entities that may be implemented as discrete or distributed components, or combined with other components, and implemented in any suitable combination and position. The various functions performed by one or more entities described herein may be implemented by hardware, firmware, and / or software. For example, some functions may be implemented by a processor that executes instructions stored in a memory.

[0025] In addition to other components not shown, the example operating environment 100 includes a network 102; a computing device 104 having a customer interface component 106; a search engine 108 having a query module 110, a generation module 112, and a search module 114; a keyword index 130; and an item database 134. It should be understood that Figure 1 The environment 100 shown is an example of a suitable operating environment. Figure 1 Each component shown in can be implemented by any type of computing device, such as the following combination Figure 8 A computing device 800 is described.

[0026] These components can communicate with each other via a network 102, which can be, but is not limited to, one or more local area networks (LANs) and / or wide area networks (WANs). In an exemplary embodiment, the network 102 includes the Internet and / or a cellular network, as well as any of a variety of possible public and / or private networks. In some aspects, the network 102 can include multiple networks, or a network of networks, but the network 102 is shown in a simpler form to avoid obscuring other aspects of the present disclosure.

[0027] It should be understood that within the scope of the present disclosure, any number of user devices, servers, and data sources may be used in the operating environment 100. Each may include a single device or multiple devices cooperating in a distributed environment. For example, the search engine 108 may be provided by multiple devices disposed in a distributed environment that together provide the functionality described herein. In addition, other components not shown may also be included in the distributed environment.

[0028] The computing device 104 can be a client device on the client side of the operating environment 100, while the search engine 108 can be on the server side of the operating environment 100. For example, the search engine 108 may include server-side software designed to work in conjunction with the client-side software on the computing device 104 to implement any combination of the features and functions discussed in this disclosure. This division of the operating environment 100 is provided in order to illustrate an example of a suitable environment, and for each implementation, it is not required that any combination of the search engine 108 and the computing device 104 remain as separate entities. Although the operating environment 100 shows a configuration in a networked environment with separate computing devices, search engines, keyword indexes, and article databases, it should be understood that other configurations in which components are combined can be adopted. For example, in some configurations, the computing device can also be used as a data source and / or can provide search capabilities.

[0029] The computing device 104 may include any type of computing device that can be used by a user. For example, in one aspect, the computing device 104 may be a Figure 8Types of computing devices 800 are described. By way of example and not limitation, the computing device may be embodied as a personal computer (PC), a laptop computer, a mobile or mobile device, a smartphone, a tablet computer, a smart watch, a wearable computer, a personal digital assistant (PDA), an MP3 player, a global positioning system (GPS) or device, a video player, a handheld communication device, a gaming device or system, an entertainment system, an in-vehicle computer system, an embedded system controller, a remote control, an appliance, a consumer electronic device, a workstation, or any combination of these devices, or any other suitable device that can perform search queries through the client interface component 106 or present notifications through the client interface component 106. A user may be associated with a computing device 104. A user may communicate with the search engine 108 through one or more computing devices, such as the computing device 104.

[0030] At a high level, the search engine 108 receives a text-based search query (e.g., a natural language query or a structured query) or an audio query including voice or other audio input from a computing device 104 (or another computing device not shown). In some aspects, the text-based query or the audio query includes one or more keywords. The search query may include any type of input from a user for initiating a search including one or more keywords. In response to receiving the search query, the search engine 108 generates and ranks text-based results in a single search result set.

[0031] In some configurations, the search engine 108 may be embodied on one or more servers. In other configurations, the search engine 108 may be implemented at least partially or entirely on a user device, such as Figure 8 The computing device 800 described in . The search engine 108 (and its components) may be embodied as a set of compiled computer instructions or functions, program modules, computer software services, or as an arrangement of processes executed on one or more computer systems.

[0032] like Figure 1As shown, the search engine 108 includes a query module 110, a generation module 112, and a search module 114. In one aspect, the functions performed by the modules of the search engine 108 are associated with one or more personal assistant applications, services, or routines. Specifically, such applications, services, or routines can be run on one or more user devices (e.g., computing device 104) or servers (e.g., search engine 108), or can be distributed across one or more user devices and servers. In some aspects, the application, service, or routine can be implemented in the cloud. In addition, in some aspects, these modules of the search engine 108 can be distributed on a network, including one or more servers and client devices (e.g., computing device 104) in the cloud, or can reside on a user device such as computing device 104.

[0033] In addition, the modules of the search engine 108 and the functions and services performed by these modules can be implemented at an appropriate abstraction layer, such as an operating system layer, an application layer, or a hardware layer. Alternatively or additionally, the functions of these modules (or various aspects of the technology described herein) can be performed at least in part by one or more hardware logic components. For example, but not limited to, illustrative types of hardware logic components that can be used include field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), etc. In addition, although the functions of specific modules shown in the search engine 108 are described herein, it is contemplated that in some aspects, the functions of one module can be shared or distributed on other modules.

[0034] The query module 110 receives a search query including one or more text-based keywords. For example, a user may enter a search query at the computing device 104 through the client interface component 106 that provides access to a search engine. As previously described, the search query may include any type of input from a user to initiate a search including one or more keywords.

[0035] The query module 110 may be configured to receive a search. The query module 110 may be configured to classify the query as an intent-based query or a product-based query. To this end, the query module 110 may employ a machine learning model (e.g., a classification model). The classification model is trained to determine whether the search query is a product-based query or an intent-based query. For example, the classification model may be trained to identify words that are characteristic of a user searching for an item to perform a task. Additionally or alternatively, the classification model may be trained to classify the search query based on knowledge learned from user feedback.

[0036] The query module 110 may also be configured to determine whether the initial search query is a cached query (whether it is a product-based query or an intent-based query). For example, as described below, the query module 110 may initially search the search log. If the initial search query is a cached query, the query module 110 may quickly retrieve search results or item lists corresponding to the task while reserving resources for other search queries that are not cached.

[0037] In addition, for example, the query module 110 can be configured to transmit the search query or the selection of items in the list of items corresponding to the task to other modules of the search engine 108, such as the generation module 112 or the search module 114. For example, if the query module 110 determines that the search query is an intent-based query, the query module 110 can transmit the search query to the generation module 112. In addition, the user query module 110 can be configured to send the search results of the search query, the list of items corresponding to the task, or the list of items associated with the search query or the task to a computing device, such as the computing device 104.

[0038] Continuing with this example, the query module 110 can cause one or more graphical user interface displays of various computing devices to display a search query, a list of items corresponding to a task, or a list of items associated with a search query or a task. In various aspects, the query module 110 causes the client interface component 106 to display a search query, a list of items corresponding to a task, or a list of items associated with a search query or a task, through which the search query is input (e.g., by a user in a search tool on a web page). In addition, the query module 110 can include an application program interface (API) that allows an application to submit a search query (and, optionally, other information, such as user information, context information, etc.) for receipt by the search engine 108.

[0039] The generation module 112 utilizes a machine learning model to generate a list of items corresponding to a task. The task may be obvious based on one or more words of the search query. In some aspects, the generation module 112 identifies or generates a task based on one or more words of the search query. The generation module 112 includes any type of generative AI model that uses a neural network to recognize patterns and structures within existing data (e.g., search queries) to generate new and original content (e.g., a list of items corresponding to a task or the task itself). Although described as a single machine learning model, the generation module 112 may be a series of machine learning models that work together to generate a list of items or a task.

[0040] The search module 114 identifies search results in response to the search query processed against the item database 134, which will be described in more detail below. For example, the search module 114 can query the keyword index 130 to identify results that meet the criteria of the search query or the item list corresponding to the task. In some aspects, the results identified in the keyword index 130 are mapped to items in the item database 134. For clarity, the items can be item listings of products and can include various additional information, such as price, price range, quality, condition, grade, material, brand, manufacturer, etc.

[0041] Search module 114 also ranks search results. In some aspects, the ranking of search results is optimized using information learned from historical search sessions or user feedback. For example, the ranking of individual items in the search results may be improved or reduced using the selections made by other users who submitted similar queries.

[0042] In some aspects, feedback may be stored in a search log. The search log may be embodied in a plurality of databases. One or more databases include one or more hardware components that are part of the search engine 108. In various aspects, the search log is configured to store information about a user's historical search session, including, for example, search queries submitted by multiple users through a client interface component (e.g., client interface component 106), a list of items corresponding to a task defined by a historical search query, search results associated with a historical search query and / or a task, a list of items in the search results, or user interactions associated with the search results (e.g., hovering, clicking, purchasing, etc.). In some embodiments, the search log stores a timestamp (e.g., day, hour, minute, second, etc.) for each user query, search result, list of items corresponding to a task, list of items associated with a search result and / or a task, user interactions with the search results and / or a list of items corresponding to a task, etc.

[0043] In addition, the information about the historical search session stored in the search log may include other result selection information, such as subsequent filters selected in response to receiving search results, item lists corresponding to tasks, and item lists. In some embodiments, the result selection information may include the time between two consecutive selections of search results, the language used by the user, and the country in which the user may be located (e.g., based on the server used to access the search engine 108). In some embodiments, other information stored associated with the historical search session may include the user's interaction with the ranking displayed within the item list, negative feedback displayed with the item list or the item list corresponding to the task, and other information, such as whether the user clicked on or viewed the document associated with the item list. User information including user cache files, cache file age, IP (Internet Protocol) address, browser user agent, etc. may also be stored in the search log. In some embodiments, user information is recorded in the search log for the entire user session or multiple user sessions.

[0044] The keyword index 130 and the item database 134 may comprise data sources or data systems that are configured to make data available to any of the various components of the operating environment 100. The keyword index 130 and the item database 134 may be discrete components separate from the search engine 108, or may be incorporated or integrated into the search engine 108 or other components of the operating environment 100. In addition, the item database 134 may store search results associated with a search query about which information may be indexed in the keyword index 130.

[0045] The keyword index 130 may take the form of an inverted index, but may also take other forms. The keyword index 130 stores information about items in a manner that allows the search engine 108 to efficiently identify search results for a search query. The search engine 108 may be configured to run any number of queries on the keyword index 130. According to an example embodiment, the keyword index 130 may include an inverted index that stores mappings from text search queries to items in the item database 134.

[0046] In practice, now turn to Figures 2 to 5 , example user interfaces 200, 300, 400, and 500 illustrate various search results. For example, Figure 2200 is an example user interface showing search results before the search system utilizes generative AI. As shown, a user of the site submits a search query 202 that includes the term "repair torn clothing." Because the search system cannot understand the user's intent, the search system returns results 204 that include items available within the e-commerce website (e.g., clothing). However, in this example, the user is searching for items for actually repairing torn clothing (e.g., ironing patches, fabric glue, sewing kits, fabric tape, etc.), and the results provided are not helpful.

[0047] refer to Figure 3 , according to one aspect of the technology described herein, the example user interface 300 shows search results after the search system utilizes generative AI. In this example, a user of the site also submitted a search query including the term "repairing torn clothes" ( Figure 3 310). However, at this point, the search system utilizes generative AI to provide a list of items corresponding to task 310. As shown, search results (e.g., ironing patches, fabric glue, sewing kits, fabric tape, etc.) 312 actually enable the user to repair torn clothes and are very helpful.

[0048] In another example, now refer to Figure 4 , example user interface 400 shows search results before the search system utilizes generative AI. As shown, a user submits a query 402 that includes the term "essentials for traveling." Because the search system cannot understand the user's intent, the search system returns results 404 of items available within the site that have titles or descriptions that match the search terms (e.g., books with terms such as "essentials," "for," and "travel" in the title or description). However, in this example, the user is searching for items needed for travel (e.g., backpacks, sleeping bags, sunscreen, cameras, etc.), and, again, the results provided are not helpful.

[0049] refer to Figure 5 , according to one aspect of the technology described herein, an example user interface 500 shows search results after the search system utilizes generative AI. In this example, the user of the site also submits a search query including the term "necessities for travel". However, at this time, the search system utilizes generative AI to provide a list of items corresponding to task 510. As shown, the search results (e.g., backpacks, sleeping bags, sunscreen, cameras, etc.) 512 actually enable the user to purchase the necessary items for the trip and are very helpful.

[0050] Figure 6 6 is a flowchart illustrating a method 600 for identifying products for completing a task using generative AI according to an aspect of the technology described herein. For example, the method 600 may be Figure 1 As shown in block 602, a search query is initially received from a user. The search query may include one or more text-based keywords. In various aspects, the search query may be performed by Figure 1 The client interface component 106 receives a search query from the computing device 104 at the search engine 108 .

[0051] In various aspects, the search query is classified as a product-based query (i.e., the user is searching for a specific product) or an intent-based query (i.e., the user is searching for an item to complete a task). Additionally or alternatively, the search query can be determined to be not a cached query. If the search query is a cached query, providing results corresponding to the cached query may be more resource-friendly than utilizing generative AI. Similarly, if the search query is a product-based query, there is no need to utilize generative AI because the user has already directed the search to a specific item or product.

[0052] At box 604, based on determining that the search query is an intent-based query, the search query is provided to the generative AI model. In some aspects, the task defined by the search query is identified by one or more components of the search engine (e.g., query module 112 or generation module 112). For example, based on one or more words of the initial search query, the task can be obvious. In some aspects, the generative AI model identifies or generates the task based on one or more words of the initial search query. The generative AI model generates a list of items corresponding to the task. Continuing with the first example above, the query can be "repairing torn clothes." The task identified by the query can be "repairing torn clothes," or the generative AI model can identify or generate the task "items for helping to repair torn clothes" based on one or more words of the initial search query. Thus, the generative AI model can generate a list of items that help to repair torn clothes (e.g., ironing patches, fabric glue, sewing kits, fabric tape, etc.).

[0053] Continuing with the second example above, the query may be "travel essentials." The task identified by the query may be "travel essentials," or the generative AI model may identify or generate the task "items useful for packing for a trip" based on one or more words of the initial search query. Thus, the generative AI model generates a list of items useful for packing for a trip (e.g., backpacks, sleeping bags, sunscreen, cameras, etc.). In various aspects, before providing the list of items to the user or performing a search for the items, the availability of the items in the list of items is initially confirmed as being available in inventory.

[0054] At box 606, a list of items corresponding to the task defined by the search query is received from the generative AI model. In various aspects, several options are provided to the user. A first option may be to perform a search for the original search query and ignore the list of items generated by the generative AI model. A second option may be to perform a search for the list of items generated by the generative AI model. A third option may be to provide the list of items corresponding to the task defined by the search query as a search suggestion to the user. In this option, a selection of a portion of the items in the list of items may be received from the user. Based on the selection, a search is performed on the portion of items. A fourth option may enable the user to add additional items to the list of items or a portion of items before performing the search. At box 608, search results are provided to the user.

[0055] Figure 7 is a flow chart illustrating a method 700 for utilizing generative artificial intelligence (AI) to identify products for completing a task according to one aspect of the technology described herein. For example, the method 700 may be performed by Figure 1 In various aspects, the search engine 108 can be used to perform Figure 1 The client interface component 106 receives a search query from the computing device 104 at the search engine 108. The search query may include one or more text-based keywords.

[0056] As shown in box 702, identify a task defined by a search query initiated by a user. If the search query is an intent-based query, all or a portion of one or more text-based keywords can be used to identify the task. Although not shown, a first machine learning model (e.g., a classification model) can be initially utilized to determine whether the search query is a product-based query or an intent-based query. In some aspects, if the initial search query is an intent-based query, it is provided to the second machine learning model. In other aspects, if the initial search query is a cached query (regardless of whether it is a product-based query or an intent-based query), it is not provided to the second machine learning model to preserve resources. Based on one or more words of the initial search query, the task can be obvious. In some aspects, a generative second machine learning model identifies or generates a task based on one or more words of the initial search query.

[0057] In box 704, a second machine learning model (e.g., a generative AI model) may be used to generate a list of items corresponding to the task. For example, assume the search query is "things to put on the dining table." The first machine learning model may determine that the search query is an intent-based query. Therefore, the intent-based query is provided to the second machine learning model. The second machine learning model may generate a list of items corresponding to "things to put on the dining table." In some aspects, the second machine learning model may first identify or generate text corresponding to the task defined by the initial search query. For example, the second machine learning may identify or generate the task "things that can be used to decorate the dining table." Based on the task, the second machine learning model may generate a list of items such as "centerpieces to put on the dining table," "candle holders," "flowers," etc. that may help the user complete the task.

[0058] In another example, assume that the search query is "materials for building a stone fountain." The first machine learning model can determine that the search query is an intent-based query. Therefore, the intent-based query is provided to the second machine learning model. The second machine learning model can generate a list of items corresponding to "materials for building a stone fountain." In some aspects, the second machine learning model can first identify or generate text corresponding to the task defined by the initial search query. For example, the second machine learning model can identify or generate the task "Materials needed to build a stone fountain." Based on the task, the second machine learning model can generate a list of items such as "fountain pool", "fountain pump", "rock accessories", etc. that can help the user complete the task.

[0059] At block 706, a search is performed on the selected portion of the list of items.

[0060] refer to Figure 8 , computing device 800 includes a bus 810 that is directly or indirectly coupled to the following devices: memory 812, one or more processors 814, one or more presentation components 816, one or more input / output (I / O) ports 818, one or more I / O components 820, and an exemplary power supply 822. Bus 810 represents one or more buses (e.g., an address bus, a data bus, or a combination thereof). Although for clarity, the bus is shown with lines. Figure 8 The blocks of the processor are, in fact, logical components, not necessarily actual components. For example, a presentation component such as a display device can be considered an I / O component. In addition, the processor has a memory. The inventors of the present disclosure recognize that this is the nature of the art and reiterate that Figure 8 The illustrations are merely illustrative of exemplary computing devices that may be used in conjunction with one or more aspects of the present technology. No distinction is made between categories such as "workstation," "server," "laptop," "handheld device," etc., as all are considered to be within the meaning of the present invention. Figure 8and reference to “computing device”.

[0061] The computing device 800 typically includes various computer-readable media. Computer-readable media can be any available media that can be accessed by the computing device 800, and includes both volatile media and non-volatile media, as well as removable media and non-removable media. By way of example and not limitation, computer-readable media can include computer storage media and communication media.

[0062] Computer storage media includes volatile and nonvolatile media, and removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by the computing device 800. Computer storage media themselves do not include signals.

[0063] Communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and includes any information delivery media. The term "modulated data signal" refers to a signal that has one or more characteristics set or changed so as to encode information in the signal. By way of example and not limitation, communication media include wired media, such as a wired network or direct wired connection, and wireless media, such as acoustic, RF, infrared and other wireless media. Any combination of the above should also be included within the scope of computer readable media.

[0064] Memory 812 includes computer storage media in the form of volatile memory and / or non-volatile memory. Memory can be removable memory, non-removable memory, or a combination thereof. Exemplary hardware devices include solid-state memory, hard disk drives, optical disk drives, etc. Computing device 800 includes one or more processors 814 that read data from various entities such as memory 812 or I / O components 820. Presentation component 816 presents data indications to a user or other device. Exemplary presentation components include display devices, speakers, printing components, vibration components, etc.

[0065] I / O ports 818 allow computing device 800 to be logically coupled to other devices, including I / O components 820, some of which may be built-in. Illustrative components include a microphone, joystick, game pad, satellite dish, scanner, printer, wireless devices, etc.

[0066] The I / O component 820 may provide a natural user interface (NUI) that processes air gestures, voice, or other physiological inputs generated by the user. In some instances, the input may be transmitted to an appropriate network element for further processing. The NUI may implement any combination of voice recognition, touch and stylus recognition, facial recognition, biometric recognition, gesture recognition on and near the screen, air gestures, head and eye tracking, and touch recognition associated with a display on the computing device 800. The computing device 800 may be equipped with a depth camera, such as a stereo camera system, an infrared camera system, an RGB camera system, and a combination thereof, for gesture detection and recognition. In addition, the computing device 800 may also be equipped with an accelerometer or gyroscope capable of detecting motion. The output of the accelerometer or gyroscope may be provided to a display of the computing device 800 to present an immersive augmented reality or virtual reality.

[0067] Some aspects of the computing device 800 may include one or more radios 824 (or similar wireless communication components). The radio 824 sends and receives radio or wireless communications. The computing device 800 may be a wireless terminal suitable for receiving communications and media through various wireless networks. The computing device 800 may communicate with other devices through wireless protocols such as code division multiple access (CDMA), global system for mobile (GSM) or time division multiple access (TDMA) and other protocols. The radio communication may be a short-range connection, a long-range connection, or a combination of short-range and long-range wireless telecommunication connections. When we refer to "short-range" and "long-range" types of connections, we do not refer to the spatial relationship between the two devices. Instead, we generally refer to short-range and long-range as different categories or types of connections (i.e., primary and secondary connections). As an example and not limitation, a short-range connection may include a Wi-Fi® connection to a device (e.g., a mobile hotspot) that provides access to a wireless communication network, such as a wireless local area network (WLAN) connection using the 802.11 protocol. A Bluetooth connection to another computing device is a second example of a short-range connection or a near-field communication connection. By way of example and not limitation, the long-range connection may include a connection using one or more of CDMA, General Packet Radio Service (GPRS), GSM, TDMA, and 802.16 protocols.

Claims

1. A method for using generative artificial intelligence (AI) to identify products for completing a task, the method comprising: receiving a search query from a user; Based on determining that the search query is an intent-based query, providing the search query to a generative AI model; receiving, from the generative AI model, a list of items corresponding to a task defined by the search query; as well as The search results are provided to the user.

2. The method according to claim 1, further comprising: providing the list of items corresponding to the task defined by the search query as search suggestions to the user; receiving a selection of a portion of items from the list of items from the user; and Based on the selection, a search is performed for the portion of items.

3. The method according to claim 1, further comprising: The search query is categorized as a product-based query or an intent-based query.

4. The method according to claim 1, further comprising: determining available items from the portion of items available in inventory; as well as A search is performed for the portion of items available in the inventory.

5. The method according to claim 1, further comprising: The user is provided with an option to perform a search against the search query or to perform a search against the list of items corresponding to the task defined by the search query.

6. The method according to claim 1, further comprising: A determination is made that the search query is not a cached query.

7. The method according to claim 1, further comprising: A task defined by the search query is identified.

8. The method according to claim 1, further comprising: The user is enabled to add additional items to the portion of items prior to performing the search.

9. One or more non-transitory computer storage media storing computer-readable instructions that, when executed by a processor, cause the processor to perform operations comprising: identifying tasks defined by user-initiated search queries; Using a machine learning model to generate a list of items corresponding to the task; as well as A search is performed on a selected portion of the list of items.

10. The one or more non-transitory computer storage media of claim 9, determining that the search query is not a cached query.

11. The one or more non-transitory computer storage media of claim 10, further comprising: The search query is categorized as a product-based query or an intent-based query.

12. The one or more non-transitory computer storage media of claim 11, further comprising: The search query is provided to the machine learning model.

13. The one or more non-transitory computer storage media of claim 9, further comprising: Available items in the selected portion of the list of items available in inventory are determined.

14. The one or more non-transitory computer storage media of claim 9, further comprising: The user is provided with an option to perform the search on the selected portion of the list of items or to add additional items to at least a portion of the list of items corresponding to the task before performing the search.

15. A system for identifying products for completing a task using generative artificial intelligence (AI), the system comprising: at least one processor; as well as One or more computer storage media storing computer readable instructions, which, when executed by the at least one processor, cause the at least one processor to perform operations, the operations comprising: receiving a search query from a user; Providing the search query to a generative AI model; receiving, from the generative AI model, a list of items corresponding to the task defined by the search query; and Search results are provided to the user, the search results including at least a portion of the list of items corresponding to the task.

16. The system of claim 15, further comprising: The list of items corresponding to the task defined by the search query is provided to the user as search suggestions.

17. The system of claim 15, further comprising: A selection of the portion of items from the list of items is received from the user.

18. The system of claim 17, further comprising: Based on the selection, a search is performed for the selection of the portion of items.

19. The system of claim 15, further comprising: The user is provided with an option to perform a search for at least a portion of the list of items corresponding to the task defined by the search query or to add additional items to the at least a portion of the list of items corresponding to the task before performing the search.

20. The system of claim 15, further comprising: The task defined by the search query is identified.

Images