Information search method and device, equipment, storage medium and program product

Abstract

The embodiment of the invention relates to an information searching method and device, equipment, a storage medium and a program product. The method comprises the steps of determining input content in response to a first interaction operation corresponding to an information search function; determining a target search result corresponding to the input content, and displaying the input content and the target search result in a search result page; the target search result comprises aggregation content matched with the input content, and the aggregation content comprises a result obtained by performing information aggregation on target text content and at least one target promotion resource; the target text content is generated by performing search task processing on the input content through a generative model; the target promotion resource is obtained by screening the plurality of initial promotion resources based on the input content through a generative model; the initial promotion resource is a resource that has a value conversion attribute and is used to promote the setting subject. In this way, the conversion rate of the target promotion resources and the satisfaction degree of the target promotion resources to user search requirements are improved.

Description

Technical Field

[0001] This disclosure relates to the field of information processing technology, and in particular to an information search method, apparatus, device, storage medium, and program product. Background Technology

[0002] With the explosive growth of online information, users typically rely on search engines to find what they need. Current information search solutions primarily involve the search engine understanding the user's search input and then performing a relevance search on various information in a database based on the understanding results to obtain the final search results. The search results obtained through this method mainly consist of web pages, occasionally including one or more promotional resources targeting specific user groups. However, these promotional resources may differ significantly from the user's search needs, not only failing to meet the user's search requirements but also significantly interfering with the user's search experience, severely impacting their search experience. Summary of the Invention

[0003] To address the aforementioned technical problems, this disclosure provides an information search method, apparatus, device, storage medium, and program product.

[0004] In a first aspect, embodiments of this disclosure provide an information search method, the method comprising: In response to the first interactive operation corresponding to the information search function, determine the input content corresponding to the first interactive operation; The system determines the target search result corresponding to the input content and displays the input content and the target search result on the search results page. The target search result includes aggregated content matching the input content, which includes results obtained by aggregating information from the target text content and at least one target promotional resource. The target text content is generated by processing the input content using a generative model. The target promotional resource is obtained by filtering multiple initial promotional resources based on the input content using the generative model. The initial promotional resource is a resource presented as a display-type information carrier, possessing value conversion attributes, and used to promote a specific target audience.

[0005] Secondly, embodiments of this disclosure also provide an information search device, the device comprising: The input content determination module is used to determine the input content corresponding to the first interactive operation in response to the information search function. The target search result display module is used to determine the target search result corresponding to the input content and display the input content and the target search result on the search result page. The target search result includes aggregated content matching the input content, and the aggregated content includes the result obtained by aggregating information from the target text content and at least one target promotional resource. The target text content is generated by processing the input content through a generative model. The target promotional resource is obtained by filtering multiple initial promotional resources based on the input content using the generative model. The initial promotional resource is a resource with value conversion attributes, presented as a display-type information carrier, and used to promote a specific target.

[0006] Thirdly, embodiments of this disclosure also provide an electronic device, the electronic device comprising: processor; Memory, used to store executable instructions; The processor is configured to read executable instructions from memory and execute the executable instructions to implement the information search method described in any embodiment of this disclosure.

[0007] Fourthly, embodiments of this disclosure also provide a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to implement the information search method described in any embodiment of this disclosure.

[0008] Fifthly, embodiments of this disclosure also provide a computer program product for executing the information search method described in any embodiment of this disclosure.

[0009] The information search method, apparatus, device, storage medium, and program product of this disclosure are capable of responding to a first interactive operation corresponding to an information search function, determining the input content corresponding to the first interactive operation; determining the target search result corresponding to the input content, and displaying the input content and the target search result on a search result page; the target search result includes aggregated content matching the input content, the aggregated content including the result obtained by aggregating information from target text content and at least one target promotional resource; the target text content is generated by performing search task processing on the input content through a generative model; the target promotional resource is obtained by filtering multiple initial promotional resources based on the input content through the generative model; the initial promotional resource is a resource with value conversion attributes, presented as a display-type information carrier, used to promote a set subject; and the system realizes the use of a number of promotional resources... The initial promotional resources in the database serve as one of the data sources for information search. Generative models are used to deeply understand and match the user's search input with the initial promotional resources, obtaining target promotional resources that are suitable for the input. Then, the target text content and target promotional resources integrated by traditional search capabilities are further integrated to obtain the final target search results. On the one hand, the search scenario is used as a new channel for promotional resource delivery, thereby increasing the exposure and conversion rate of promotional resources. On the other hand, the powerful natural language processing capabilities of large-scale generative models greatly improve the relevance between the target promotional resources obtained from the search and the input content. This ensures that the target promotional resources meet the strong experiential constraints of the search scenario in terms of relevance and quality, thus greatly reducing the interference of target promotional resources on users, improving the degree to which target promotional resources meet users' search needs, and ultimately enhancing the user's search experience.

[0010] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in the embodiments of this disclosure are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse. Attached Figure Description

[0011] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0012] Figure 1A flowchart illustrating an information search method provided in an embodiment of this disclosure; Figure 2 This is a schematic diagram illustrating an information search result corresponding to an addressing search category provided in an embodiment of this disclosure; Figure 3 This is a schematic diagram illustrating information search results corresponding to a consumer search category provided in an embodiment of this disclosure; Figure 4 This is a schematic diagram illustrating information search results corresponding to a content search category provided in an embodiment of this disclosure; Figure 5 This is a schematic diagram illustrating multi-terminal data interaction of an information search method provided in an embodiment of the present disclosure; Figure 6 This is a schematic diagram of the structure of an information search device provided in an embodiment of the present disclosure; Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0013] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0014] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0015] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0016] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0017] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0018] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0019] Promotional resources are valuable resources used to promote specific entities (such as brands, products, or services), and they differ significantly from most information on the internet. Therefore, in information search scenarios using relevant technologies, promotional resources are rarely recalled, making it difficult to effectively deploy and convert them into value. The reasons for this are mainly twofold: On the one hand, given the significant difference between promotional resources and most information on the internet, two types of search engines exist: traditional search engines targeting general information and resource search engines targeting promotional resources. These two types differ significantly in their interpretation of search intent, search objects, search filtering, and ranking. Traditional search engines tend to interpret search intent based on information needs; for example, a search for "how to make a cake" might be interpreted as a learning need. Resource search engines, on the other hand, tend to interpret search intent based on value conversion needs; for example, a search for "how to make a cake" might be interpreted as a need to acquire baking tools or ingredients. Traditional search engines search for webpage databases, while resource search engines search for promotional resource databases. Traditional search engine filtering and ranking primarily focus on the relevance and authority of webpages, while resource search engine filtering and ranking primarily focus on the bidding price, conversion rate, and cost / benefit of promotional resources. Given the aforementioned differences, in information search scenarios using related technologies, traditional search engines lack the ability to recall promotional resources. Even if a small number of promotional resources are forcibly recalled from the promotional resource database, the degree of matching between these resources and the search requirements is low, thereby interfering with the user's information acquisition process and severely impacting the user's search experience.

[0020] On the other hand, although the promotional resources released by the providers of promotional resources may contain a lot of high-value information, the channels for setting up and distributing promotional resources usually do not include search scenarios. Therefore, it is difficult to recall promotional resources in the information search scenarios of related technologies.

[0021] Based on the above, this disclosure provides a technical solution for information search to break through the barriers between traditional information search and promotional resource search, enabling traditional search engines to access promotional resource databases for searching. Furthermore, it adds processing of search experience constraints such as relevance and quality to the search logic for promotional resources, allowing search engines to recall target promotional resources and make them better match the search needs of the input content, thereby expanding the distribution channels of promotional resources and improving the exposure and conversion rate of promotional resources.

[0022] The information search method provided in this disclosure is applicable to search scenarios. This method can be executed by an information search device, which can be implemented in software and / or hardware, and can be integrated into an electronic device with search functionality and certain data processing capabilities. This electronic device may include, but is not limited to, smartphones, personal digital assistants (PDAs), tablet computers (Tablet PCs), laptops, mobile workstations, desktop computers, or servers.

[0023] Figure 1 A flowchart illustrating an information search method provided in an embodiment of this disclosure is shown. Figure 1 As shown, this information search method may include the following steps: S110. In response to the first interactive operation corresponding to the information search function, determine the input content corresponding to the first interactive operation.

[0024] The first interactive operation is the one that triggers the launch of the information search function. For example, if the launch entry point for the information search function is a traditional search box, then the first interactive operation could be an input operation on the search box; if the launch entry point for the information search function is a human-computer dialogue interface, then the first interactive operation could be an input operation on the dialogue content input box in the human-computer dialogue interface; if the launch entry point for the information search function is voice wake-up, photo wake-up, gesture wake-up, etc. (such as the various wake-up methods of a mobile assistant), then the first interactive operation could correspond to voice interaction, photo interaction, gesture interaction, etc.

[0025] Specifically, when a user has an information search need, they can perform a first interactive operation on the electronic device that carries the client for the information search function, according to the activation / wake-up method of the information search function. In response to this first interactive operation, the electronic device can obtain its corresponding input content, such as search terms / search statements entered in text form, dialogue content entered in text form, results obtained by converting speech to text from content entered in voice form, results obtained by performing image recognition and conversion from content entered in image form, or preset text content corresponding to a gesture. This input content serves as the user's search request.

[0026] S120. Determine the target search result corresponding to the input content, and display the input content and the target search result on the search results page; the target search result includes aggregated content that matches the input content, and the aggregated content includes the result obtained by aggregating information from the target text content and at least one target promotional resource; the target text content is generated by processing the input content through a generative model; the target promotional resource is obtained by filtering multiple initial promotional resources based on the input content through a generative model.

[0027] The target search results include the target promotional resources. Target promotional resources are those retrieved from the initial promotional resources in the promotional resource database. Initial promotional resources are resources presented in a display-type information carrier (such as text, images, or videos), possess value conversion attributes, and are used to promote the designated subject. The search results page is the page used to display the search results, and its implementation is adapted to the first interactive operation. For example, if the first interactive operation is inputting into a search box, the search results page is an information page containing multiple search results; if the first interactive operation is inputting into an input box in a dialog interface, the search results page is a dialog interface; if the first interactive operation is activating various information carriers, the search results page could be a corresponding pop-up page or a floating window, etc.

[0028] Specifically, after receiving input content, the electronic device can invoke the independently encapsulated functional module corresponding to the information search function based on the input content. If the functional module is integrated locally, the electronic device can directly trigger the functional module to execute the subsequent information search process. If the functional module is integrated on the server side, the electronic device can use the input content as basic data to construct a remote call request corresponding to the functional module and send it to the server so that the server can execute the subsequent information search process.

[0029] The general approach to information search can be divided into two branches. One branch follows the traditional search logic of a search engine. It utilizes generative models to process the input content through content understanding and keyword extraction. The results are then used to enhance the retrieval of web pages in a database, recalling at least one web page matching the input content (called the target web page). Finally, the content of each target web page is subjected to in-depth analysis and information integration to generate the target text content. The other branch uses a promotional resource database as the search object. Generative models are used to pre-understand the content of each initial promotional resource in the database. The results are used as the resource understanding results for the initial promotional resources. Then, a two-sided matching process is performed between the content understanding results and the resource understanding results to recall at least one target promotional resource matching the input content. Finally, the target text content and target promotional resources obtained from both branches are aggregated to generate the final target search results.

[0030] The target search results not only include target webpage data obtained from the search logic of traditional search engines, but also target promotional resources obtained from the search logic of resource search engines. Moreover, both the target webpage data and the target promotional resources meet the strong experience constraints of traditional search, such as relevance and authority. Furthermore, both are subjected to in-depth analysis and information integration through generative models, resulting in target search results with good readability and high information acquisition efficiency.

[0031] Subsequently, the electronic device obtains the target search results from the local device or the server, and displays the target search results on the search results page in a display-type information carrier style (such as information list, information card, etc.) with a high degree of visualization and easy interaction, according to the content type contained in the target search results.

[0032] It should be noted that, as Figure 2 As shown, in addition to displaying the input content 210 and the target search result 220, the search results page (such as the dialog interface corresponding to the information search function) can also display web page data retrieved from the web page database by the search logic of traditional search engines, i.e., other search results 230, in order to further improve the information comprehensiveness and information acquisition efficiency of the search scenario.

[0033] In related technologies, generative models are prone to misidentification when identifying search needs / search purposes from input content, especially for long-tail content or dialogue scenarios with many rounds. Generative models may fail to accurately identify the user's correct search needs, leading to misunderstandings of the input content and consequently inaccurate search retrieval results. Therefore, this embodiment divides initial promotional resources into three main categories based on their value, quality, and quantity: address-based promotional resources, consumer-based promotional resources, and content-based promotional resources. Correspondingly, search needs can also be categorized into address-based search needs (i.e., address-based search category), consumer-based search needs (i.e., consumer search category), and content-based search needs (i.e., content search category). This categorization of search needs narrows the semantic scope during input content identification, reduces ambiguity in content understanding, and improves the accuracy of search need comprehension, providing a solid data foundation for accurate retrieval of subsequent promotional resources.

[0034] In some embodiments, if the target promotional resource corresponds to the address search category, and the target promotional resource includes platform promotional resources and / or application promotional resources corresponding to the set subject, then the target search result is displayed on the search results page, including: displaying the target search result in the form of a graphic carrier and the resource access entry control corresponding to the target search result on the search results page.

[0035] Among them, address-based search refers to the type of search request expressed by users through search behavior, aiming to obtain the spatial location information (such as physical address) or logical orientation information (such as network address, platform location path) of the target object. The target object here can correspond to the defined subject in the promotional resources, which includes, but is not limited to, physical locations (such as offline stores), online resources (such as official websites), and digital content carriers (such as mini-programs, applications, etc.). Platform promotional resources refer to resources that promote the network address carrying the online resource, such as web pages or videos promoting a travel platform. Application promotional resources refer to resources that promote digital content carriers, such as web pages or videos promoting an application / mini-program that provides travel-related services or products. Resource access entry controls are interactive controls used to trigger access to the target promotional resource.

[0036] Specifically, electronic devices can utilize generative models to perform deep identification and classification of user input regarding search needs. See also... Figure 2For the input content 210, "providing websites or platforms for booking vacation islands," the electronic device can identify that it belongs to the address search category. Then, following the aforementioned information search process, it can determine the appropriate target promotional resources, including platform promotional resources and / or application promotional resources. The target search results contain a small amount of target text content that briefly describes the target promotional resources, while a larger portion corresponds to the target promotional resources themselves. Thus, on the search results page (such as...) Figure 2 The dialog interface shown can display the target text content from the search results in text format, and the introduction content of the platform / application / mini-program in the target promotional resources in image format (such as the name and introduction in text format, and the introduction content in image format, etc.). Furthermore, resource access controls are displayed around each target promotional resource, such as the "Open" button 221 for entering the website landing page for platform promotional resources, the "Start" button for launching the mini-program for mini-program promotional resources, the "Start" button for launching the application if the application corresponding to the application promotional resource is already installed, and the "Install" button 222 for installing the application if the application corresponding to the application promotional resource is not installed, etc. This not only effectively reveals the introductory information of the target promotional resources for the address search category, increasing the exposure and conversion rate of the target promotional resources, but also provides a more convenient interactive path for subsequent in-depth conversion operations for the platform / application / mini-program, simplifying the interaction path from first entering the search result landing page to entering the platform / application / mini-program landing page, thereby further improving the satisfaction rate of users' address search needs.

[0037] In other embodiments, if the target promotional resource corresponds to a consumer search category, and the target promotional resource includes promotional resources of the setting subject under the consumer scenario adaptation carrier, then the target search results are displayed on the search results page, including: displaying the target text content in the form of a text carrier and displaying the target promotional resource in the form of an information card with an embedded resource access entry on the search results page.

[0038] Among them, the consumer search category refers to the type of search demand expressed by users through search behavior, aiming to obtain and confirm the corresponding carrier (such as target item unit, geographically related service, collaborative consumption demand aggregation service) to achieve resource exchange. Target item unit refers to a tangible object with a physical or virtual form (such as e-commerce goods, digital memberships, etc.). Geographically related services refer to service execution forms based on geographical location (such as local food delivery, housekeeping services, beauty services, etc.). Collaborative demand aggregation services refer to a batch demand fulfillment mode based on the aggregation of similar demands from multiple users (such as group buying of goods, service group buying, etc.). Consumption scenario adaptation carrier refers to the technical carrier form that matches the user's consumption demand scenario, including but not limited to e-commerce consumption scenario carriers (such as online shopping platforms), geographically related service consumption scenario carriers (such as local service corresponding platforms), and collaborative demand aggregation service consumption scenario carriers (such as group buying platforms), etc.

[0039] Specifically, electronic devices can utilize generative models to perform deep identification and classification of user input regarding search needs. See also Figure 3 For the input content 310, "Recommendations for effective moisturizing face creams in winter," the electronic device can identify that it belongs to the consumer search category. Then, following the aforementioned information search process, it can determine the appropriate target promotional resources, including promotional resources for the specified subject within the appropriate consumer scenario. For example, if the specified subject is "face cream of brand xx," then the target promotional resources could be promotional web pages or videos of brand xx's face cream on e-commerce platforms / local service platforms / group buying platforms. The target search results then include introductory content about the specified subject extracted from the target text content and target promotional resources, as well as access points to the target promotional resources. Figure 3 As shown, in the search results page presented in a dialog interface, the electronic device displays the input content 310 and the target search results 320. The target search results 320 display introductory content 321 of the aforementioned target entity in text format, and the target promotional resource is displayed as an information card (such as a product card 322) with an embedded resource access portal. This not only satisfies the information search needs of the input content through the aforementioned introductory content, increasing the exposure and conversion rate of the target promotional resource, but also directly reveals the consumer interaction card of the target promotional resource, better assisting users in diversified information consumption and improving the acquisition efficiency of the target entity corresponding to the target promotional resource.

[0040] In some other embodiments, if the target promotion resource corresponds to a content search category, and the target promotion resource includes multimodal knowledge promotion resources provided by the setting entity and constructed based on industry expertise, then the target search results are displayed on the search results page, including: displaying the target search results and the setting entity identifier corresponding to the target search results in the form of graphic carriers on the search results page.

[0041] Content search categories refer to the types of search needs expressed by users through search behavior, aimed at obtaining informational content to satisfy cognitive needs (such as answering questions, learning knowledge, acquiring information, understanding principles, etc.). Industry expertise refers to precise and professional information within an industry provided by the target entity. Multimodal knowledge promotion resources refer to promotional resources provided by the target entity (such as brands, industry organizations, etc.), centered on industry expertise, and constructed through various technical forms (multimodal) such as text, audio / video, and thematic integration. Target entity identifiers are identifying information about the target entity, such as text tags or buttons embedding the logical access address corresponding to the target entity.

[0042] Specifically, electronic devices can utilize generative models to perform in-depth identification and classification of user input content for search needs. For example, the generative model can employ exclusion rules to identify the search category of the input content as either address search or consumption search, thus classifying it as a content search category; and / or, the generative model can match the input content with core keywords (such as information answers, knowledge acquisition, tutorials, learning, news, etc.) or core scenarios (information-related needs scenarios such as knowledge answers, information acquisition, method tutorials, and opinion discussions) corresponding to the content search category, and confirm it as a content search category when a match is successful.

[0043] See Figure 4For the input content 410, "What is the application-to-admission ratio of xx University's xxxx College?", the electronic device can identify that it belongs to the content search category using the aforementioned method. Then, following the aforementioned information search process, it can determine the suitable target promotional resources, including multimodal knowledge promotional resources provided by the setting entity and built based on industry expertise. In the search results page presented in a dialog interface, the electronic device displays the input content 410 and the target search result 420. The target search result 420 displays the answer to the input content 421 in text format and the target promotional resource 422 in image and text format. Based on this, the electronic device can display the setting entity's identifier 423 in the surrounding area of ​​the target promotional resource 422. This not only enhances the authority of the search results and its satisfaction rate with search needs through the industry expertise within the target promotional resources provided by the setting entity, but also improves the exposure and conversion rate of the target promotional resources. Additionally, the setting entity's identifier can be used to attach relevant leads to the setting entity, increasing incremental conversions.

[0044] The information search method provided in this disclosure is capable of responding to a first interactive operation corresponding to an information search function, determining the input content corresponding to the first interactive operation; determining the target search result corresponding to the input content, and displaying the input content and the target search result on the search result page; the target search result includes aggregated content matching the input content, and the aggregated content includes the result obtained by aggregating information from the target text content and at least one target promotional resource; the target text content is generated by performing search task processing on the input content through a generative model; the target promotional resource is obtained by filtering multiple initial promotional resources based on the input content through a generative model; the initial promotional resource is a resource with value conversion attributes, presented in the form of a display-type information carrier, used for promoting a specific subject; and realizes the use of each initial promotional resource in the promotional resource database as a data source for information search. Based on one source, and utilizing generative models to deeply understand and match the user's search input with the initial promotional resources, this approach obtains target promotional resources that are adapted to the input. Then, it further integrates the target text content and target promotional resources integrated by traditional search capabilities to obtain the final target search results. On one hand, it uses the search scenario as a new channel for promotional resource delivery, thereby increasing the exposure and conversion rate of promotional resources. On the other hand, leveraging the powerful natural language processing capabilities of large-scale generative models, it significantly improves the relevance between the target promotional resources obtained from the search and the input content. This ensures that the target promotional resources meet the strong experiential constraints of the search scenario in terms of relevance and quality, thus greatly reducing the interference of target promotional resources on users, improving the degree to which target promotional resources meet users' search needs, and ultimately enhancing the user's search experience.

[0045] In some embodiments, after S120, the method further includes: in response to a second interactive operation on the target search result, triggering the display of the operation result corresponding to the second interactive operation on the target page corresponding to the target promotion resource, and triggering the execution of value conversion processing according to the value conversion method corresponding to the target promotion resource.

[0046] The second interactive operation is an interactive operation for the in-depth conversion of the target promotional resources, which is adapted to the resource type of the target promotional resources.

[0047] Specifically, see Figure 5 After the target search result is displayed on the client's electronic device, the user can perform further interactive operations according to the interaction method corresponding to the target promotional resource displayed on the search results page, i.e., perform a second interactive operation. The electronic device then responds to this second interactive operation by executing the corresponding processing logic to trigger the display of the target page and show the operation result 512 within it. For example, for... Figure 2 The target promotional resource shown includes a resource access control. Users can trigger the "Open" button 221 to display the corresponding platform's landing page (as the operation result) for detailed information on the search results page or a new page (as the target page). Alternatively, users can trigger the "Install" button 222 to display the corresponding application's download page on the search results page or a new page (as the target page) and automatically execute the installation process to complete the application's download and installation (as the operation result). For example, for... Figure 3 The target promotional resource shown allows users to trigger the corresponding product card 322, which in turn displays the product details (as the operation result) on the search results page or a new page (as the target page), providing an interactive page for product acquisition and confirmation. For example, for... Figure 4 The target promotional resources shown can be accessed by users through triggering the target promotional resource or its corresponding designated entity identifier 423. This will display the landing page corresponding to the target promotional resource or the platform landing page of the designated entity (as the operation result) on the search results page or a new page (as the target page) for detailed information. Through the above-mentioned in-depth interaction, in-depth conversion of the target promotional resources can be achieved, triggering value conversion processing (such as statistical analysis of relevant indicators and value calculation) according to the value conversion method corresponding to the target promotional resources.

[0048] In some embodiments, taking the above information search process executed on the server as an example, it may specifically include the following steps A to D to obtain the target search results.

[0049] Step A: Use a generative model to perform content understanding and search category classification on the input content, and determine the content understanding result of the input content and the target search category to which the input content belongs.

[0050] The target search category can be either address search, consumer search, or content search.

[0051] Specifically, see Figure 5 The user performs the first interactive operation on the electronic device where the client is located, and the electronic device can obtain input content 501. Then, the electronic device can generate a remote call request based on the input content 501 and send it to the server so that the server can obtain the input content and execute the specific information search process internally.

[0052] The server first uses a generative model to perform content understanding and search category classification on the input content, and obtains the content understanding result 502 that can be interpreted by the generative model and the target search category 503 to which it belongs.

[0053] It should be noted that if the input content is dialogue content, the server can obtain the dialogue content of the current round and its context dialogue content in order to gain a deeper and more accurate understanding and classification of the input content.

[0054] It should also be noted that if the search category classification fails, that is, if the generative model determines that the search request of the input content does not belong to the address search category, consumption search category, or content search category, then it is considered that the current dialogue progress has not triggered the search process, and the search process may not be executed until the search request is classified into one of the above categories, then the subsequent steps will continue.

[0055] Step B: Using a generative model, based on the content understanding results, filter and process multiple web page data to determine at least one target web page data, and integrate the information of each target web page data to generate target text content.

[0056] Specifically, the server can utilize a generative model to perform enhanced search processing on the webpage database. This involves matching and filtering the content understanding result 502 of the input content with the data from various webpages to obtain at least one target webpage that is highly relevant to the input content. Then, in-depth analysis and information integration are performed on each target webpage to obtain the target text content 504, which serves as a fallback search result.

[0057] Step C: Based on the target search category, content understanding results, and resource understanding results corresponding to each initial promotional resource, filter each initial promotional resource to determine the target promotional resource.

[0058] The resource understanding results are obtained in advance by using a generative model to understand the content of each initial promotional resource.

[0059] Specifically, see [link to relevant documentation] Figure 5 The server can invoke a generative model to conduct in-depth analysis and understanding of each initial promotional resource in the promotional resource database 505, based on four dimensions: industry, category, target audience, and display format. This yields a resource understanding result 506 for each initial promotional resource that can be interpreted by the generative model. Then, the server can perform matching and filtering processes on each initial promotional resource based on the target search category 503, the content understanding result 502, and the resource understanding result 506, ultimately obtaining the target promotional resource.

[0060] In some embodiments, step C specifically includes: performing category matching based on the target search category and the resource categories in the resource understanding results corresponding to each initial candidate resource, and selecting multiple intermediate promotional resources from each initial promotional resource; performing relevance matching based on the content understanding results and the remaining understanding results in the resource understanding results corresponding to each intermediate promotional resource, and selecting at least one candidate promotional resource from each intermediate promotional resource; and performing multi-dimensional sorting and filtering based on the content understanding results and the remaining understanding results in the resource understanding results corresponding to each candidate promotional resource, and selecting at least one target promotional resource from each candidate promotional resource.

[0061] The remaining understanding results are those other than resource categories within the resource understanding results. The multi-dimensional approach includes at least relevance, quality, and resource delivery effectiveness metrics. Resource delivery effectiveness metrics can include cost-related metrics (such as bid), conversion efficiency metrics (such as conversion rate), and value conversion metrics (such as revenue or investment cost).

[0062] Specifically, see [link to relevant documentation] Figure 5 The server-side process of recalling target promotional resources can specifically include: the server can first use the target search category to perform search category matching with the resource categories in the resource understanding results corresponding to each initial candidate resource, so as to filter out multiple initial promotional resources that match the target search category from all initial promotional resources, as intermediate promotional resources 507. For example, if the target search category is the address search category, then each intermediate promotional resource belongs to platform promotional resources, application promotional resources, and mini-program promotional resources, etc., but does not include consumer-related promotional resources and content-related promotional resources.

[0063] Then, the server uses the content understanding result 502 and the remaining understanding results in the resource understanding results corresponding to each intermediate promotion resource 507 to perform coarse-grained relevance matching, so as to broadly recall multiple intermediate promotion resources that are relevant to the input content from each intermediate promotion resource 507 as candidate promotion resources 508.

[0064] Next, the server can use the remaining understanding results from the content understanding results 502 and the resource understanding results corresponding to each candidate promotion resource 508 to perform fine-grained relevance matching, quality screening, and resource delivery effect screening. This allows for the refined selection of at least one candidate promotion resource from each candidate promotion resource 508, which will be designated as the target promotion resource 509. Then, the server can sort each target promotion resource 509 according to multiple dimensions, including relevance, quality, and resource package delivery effect, to obtain the final recall results for the promotion resources.

[0065] In some embodiments, the above-mentioned correlation matching based on the content understanding results and the remaining understanding results in the resource understanding results corresponding to each intermediate promotional resource, and screening at least one candidate promotional resource from each intermediate promotional resource, includes: performing correlation matching based on the content understanding results, user feature data, and the remaining understanding results in the resource understanding results corresponding to each intermediate promotional resource, and screening at least one candidate promotional resource from each intermediate promotional resource.

[0066] User feature data is a set of non-identifiable features obtained by anonymizing, de-identifying, de-sensitizing, and aggregating user behavior data. It may include at least one of user search behavior features, demand preference features, and search scenario adaptation features.

[0067] Specifically, in order to further improve the recall accuracy of promotional resources, this embodiment can further introduce user feature data to enhance the personalization of target promotional resources, thereby further improving the satisfaction rate of target promotional resources with user search needs.

[0068] In practice, the server can obtain user characteristic data 510 through various channels and use it as input data for screening candidate promotional resources. This allows for further filtering out intermediate promotional resources that are irrelevant to the user, improving the accuracy of candidate promotional resources, reducing invalid filtering, and further enhancing the search efficiency of promotional resources.

[0069] In one example, the above-mentioned correlation matching based on the content understanding results, user feature data, and the remaining understanding results in the resource understanding results corresponding to each intermediate promotional resource, to select at least one candidate promotional resource from each intermediate promotional resource includes: performing user dimension matching based on the user feature data and the remaining understanding results in the resource understanding results corresponding to each intermediate promotional resource, to select at least one candidate promotional resource from each intermediate promotional resource; and performing correlation matching based on the content understanding results and the remaining understanding results in the resource understanding results corresponding to each intermediate promotional resource, to select at least one candidate promotional resource from each intermediate promotional resource.

[0070] Specifically, in this example, the server can process user feature data and content understanding results in parallel as different recall channels. This allows the server to recall candidate promotional resources that have a coarse-grained relevance to the input content, as well as candidate promotional resources that match user features. Finally, these candidate promotional resources are merged to obtain the final candidate promotional resources.

[0071] In another example, the above-mentioned correlation matching based on the content understanding results, user feature data, and the remaining understanding results in the resource understanding results corresponding to each intermediate promotional resource, and the selection of at least one candidate promotional resource from each intermediate promotional resource includes: constructing comprehensive feature data using the content understanding results and user feature data, and performing correlation matching based on the comprehensive feature data and the remaining understanding results in the resource understanding results corresponding to each intermediate promotional resource, and selecting at least one candidate promotional resource from each intermediate promotional resource.

[0072] Specifically, in this example, the server first merges user feature data and content understanding results to obtain a comprehensive feature vector (i.e., comprehensive feature data), and then uses this comprehensive feature data to perform coarse-grained relevance matching on each intermediate promotion resource to obtain each candidate promotion resource.

[0073] It should be noted that the aforementioned user characteristic data can also be used in the feedback iteration process. For example... Figure 5 As shown, the client can send data such as the second interactive operation performed by the user and its corresponding result 512 back to the server. The server can then analyze this feedback data to dynamically optimize user characteristic data, thereby continuously optimizing the selection logic of promotional resources and further improving the accuracy of target promotional resources.

[0074] Step D: Using a generative model, integrate the target text content and target promotional resources to generate target search results.

[0075] Specifically, the server can then use a generative model to perform in-depth analysis and information integration on the target text content 504 and each target promotional resource 509 to generate the target search result 511. This target search result 511 can be fed back to the client by the server so that the client can display and process it.

[0076] The following are embodiments of the information search device provided in this disclosure. This device and the information search methods in the above embodiments belong to the same inventive concept. For details not described in detail in the embodiments of the information search device, please refer to the embodiments of the above information search methods.

[0077] Figure 6 A schematic diagram of the structure of an information search device provided in an embodiment of this disclosure is shown. Figure 6 As shown, the information search device 600 may include: The input content determination module 610 is used to determine the input content corresponding to the first interactive operation in response to the information search function. The target search result display module 620 is used to determine the target search result corresponding to the input content and display the input content and the target search result on the search result page. The target search result includes aggregated content matching the input content, which includes the result obtained by aggregating information from the target text content and at least one target promotional resource. The target text content is generated by processing the input content through a generative model. The target promotional resource is obtained by filtering multiple initial promotional resources based on the input content using a generative model. The initial promotional resource is a resource with value conversion attributes, presented as a display-type information carrier, and used to promote the designated subject.

[0078] The information search device provided in this disclosure uses each initial promotional resource in the promotional resource database as one of the data sources for information search. It utilizes a generative model to deeply understand and match the user's search input with each initial promotional resource to obtain target promotional resources that match the input. Then, it further integrates the target text content and target promotional resources integrated by traditional search capabilities to obtain the final target search results. On one hand, it uses the search scenario as a new channel for promoting resources, thereby increasing the exposure and conversion rate of promotional resources. On the other hand, by leveraging the powerful natural language processing capabilities of a large-scale generative model, it greatly improves the relevance between the target promotional resources obtained from the search and the input content. This ensures that the target promotional resources meet the strong experiential constraints of the search scenario in terms of relevance and quality, thereby significantly reducing the interference of target promotional resources on users, improving the degree to which target promotional resources meet users' search needs, and ultimately enhancing the user's search experience.

[0079] In some embodiments, the information search device 600 further includes an operation result display module, used for: After displaying the input content and target search results on the search results page, in response to the second interactive operation on the target search results, the operation result corresponding to the second interactive operation is displayed on the target page corresponding to the target promotion resource, and the value conversion processing is executed according to the value conversion method corresponding to the target promotion resource.

[0080] In some embodiments, the target search result display module 620 is specifically used for: If the target promotional resource corresponds to the address search category, and the target promotional resource includes the platform promotional resource and / or application promotional resource corresponding to the set subject, then the target search result and the resource access entry control corresponding to the target search result will be displayed in the search results page in the form of graphic carrier. If the target promotional resource corresponds to the consumer search category, and the target promotional resource includes the promotional resources of the set subject under the consumer scenario adaptation carrier, then in the search results page, the target text content is displayed in the form of a text carrier, and the target promotional resource is displayed in the form of an information card with an embedded resource access entry. If the target promotional resource corresponds to a content search category, and the target promotional resource includes multimodal knowledge promotional resources provided by the setting entity and built based on industry expertise, then the target search results and the setting entity identifier corresponding to the target search results will be displayed in the search results page in the form of images and text.

[0081] In some embodiments, the information search device 600 further includes a target search result determination module, comprising: The target search category determination submodule is used to perform content understanding and search category classification on the input content using a generative model, and to determine the content understanding result of the input content and the target search category to which the input content belongs; wherein, the target search category is address search category, consumption search category or content search category; The target text content generation submodule is used to use a generative model to filter and process multiple web page data based on the content understanding results, determine at least one target web page data, and integrate the information of each target web page data to generate target text content. The target promotion resource determination submodule is used to filter each initial promotion resource based on the target search category, content understanding results, and resource understanding results corresponding to each initial promotion resource, and to determine the target promotion resource; wherein, the resource understanding results are obtained in advance by using a generative model to perform content understanding on each initial promotion resource; The target search result generation submodule is used to integrate information from target text content and target promotional resources using a generative model to generate target search results.

[0082] In some embodiments, the target promotion resource determination submodule is specifically used for: Based on the target search category and the resource category in the resource understanding results corresponding to each initial candidate resource, category matching is performed to filter out multiple intermediate promotion resources from each initial promotion resource; Based on the content understanding results and the remaining understanding results in the resource understanding results corresponding to each intermediate promotion resource, a relevance matching is performed to select at least one candidate promotion resource from each intermediate promotion resource. Based on the content understanding results and the remaining understanding results in the resource understanding results corresponding to each candidate promotion resource, a multi-dimensional sorting and filtering is performed to select at least one target promotion resource from each candidate promotion resource; among which, the multi-dimensional factors include at least the relevance dimension, the quality dimension, and the resource placement effect indicator dimension.

[0083] In some embodiments, the target promotion resource determination submodule is further specifically used for: Based on the content understanding results, user feature data, and the remaining understanding results in the resource understanding results corresponding to each intermediate promotional resource, relevance matching is performed to select at least one candidate promotional resource from each intermediate promotional resource.

[0084] The information search device provided in this disclosure can execute the information search method provided in any embodiment of this disclosure, and has the corresponding functional modules and beneficial effects of executing the method.

[0085] It is worth noting that in the embodiments of the above information search device, the various modules and sub-modules are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional module / sub-module are only for easy differentiation and are not used to limit the scope of protection of this disclosure.

[0086] This disclosure also provides an electronic device that may include a processor and a memory, the memory being used to store executable instructions. The processor can be used to read the executable instructions from the memory and execute the executable instructions to implement the information search method described in the above embodiments.

[0087] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure is shown.

[0088] like Figure 7As shown, the electronic device 700 may include a processing unit 701 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 702 or a program loaded from storage device 708 into random access memory (RAM) 703. The RAM 703 also stores various programs and data required for the operation of the electronic device 700. The processing unit 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output interface (I / O interface) 705 is also connected to the bus 704.

[0089] Typically, the following devices can be connected to I / O interface 705: input devices 706 including, for example, touch screens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 707 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 708 including, for example, magnetic tapes, hard disks, etc.; and communication devices 709. Communication device 709 allows electronic device 700 to communicate wirelessly or wiredly with other devices to exchange data.

[0090] It should be noted that, Figure 7 The illustrated electronic device 700 is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein. That is, although... Figure 7 An electronic device 700 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0091] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 709, or installed from storage device 708, or installed from ROM 702. When the computer program is executed by processing device 701, it performs the functions defined in the information search method of any embodiment of this disclosure.

[0092] This disclosure also provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to implement the information search method in any embodiment of this disclosure.

[0093] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. Computer-readable storage media can be, for example, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. The transmitted data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, radio frequency (RF), etc., or any suitable combination thereof.

[0094] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol, such as Hypertext Transfer Protocol (HTTP), and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include Local Area Networks (LANs), Wide Area Networks (WANs), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0095] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0096] The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the information search method described in any embodiment of this disclosure.

[0097] In embodiments of this disclosure, computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include, but are not limited to, object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0098] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of devices, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0099] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Array (FPGA), Application Specific Integrated Circuit (ASIC), Application Specific Standard Parts (ASSP), System on Chip (SOC), Complex Programmable Logic Device (CPLD), and so on.

[0100] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0101] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0102] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. An information search method, characterized in that, include: In response to the first interactive operation corresponding to the information search function, determine the input content corresponding to the first interactive operation; The system determines the target search result corresponding to the input content and displays the input content and the target search result on the search results page. The target search result includes aggregated content matching the input content, which includes results obtained by aggregating information from the target text content and at least one target promotional resource. The target text content is generated by processing the input content using a generative model. The target promotional resource is obtained by filtering multiple initial promotional resources based on the input content using the generative model. The initial promotional resource is a resource presented as a display-type information carrier, possessing value conversion attributes, and used to promote a specific target audience.

2. The method according to claim 1, characterized in that, After displaying the input content and the target search result on the search results page, the method further includes: In response to the second interactive operation on the target search result, the operation result corresponding to the second interactive operation is displayed on the target page corresponding to the target promotion resource, and value conversion processing is executed according to the value conversion method corresponding to the target promotion resource.

3. The method according to claim 1, characterized in that, The target search results are displayed on the search results page, including: If the target promotional resource corresponds to the address search category, and the target promotional resource includes the platform promotional resources and / or application promotional resources corresponding to the set subject, then the target search result and the resource access entry control corresponding to the target search result are displayed in the search results page in the form of graphic carrier. If the target promotional resource corresponds to a consumer search category, and the target promotional resource includes the promotional resources of the set subject under the consumer scenario adaptation carrier, then in the search results page, the target text content is displayed in the form of a text carrier, and the target promotional resource is displayed in the form of an information card with an embedded resource access entry. If the target promotion resource corresponds to a content search category, and the target promotion resource includes multimodal knowledge promotion resources provided by the designated entity and constructed based on industry expertise, then the target search result and the designated entity identifier corresponding to the target search result will be displayed in the search results page in the form of a graphic carrier.

4. The method according to any one of claims 1 to 3, characterized in that, The target search result is determined in the following way: The generative model is used to perform content understanding and search category classification on the input content to determine the content understanding result of the input content and the target search category to which the input content belongs; wherein, the target search category is an address search category, a consumption search category, or a content search category; Using the generative model, multiple web page data are filtered based on the content understanding results to determine at least one target web page data, and information is integrated from each target web page data to generate the target text content; Based on the target search category, the content understanding results, and the resource understanding results corresponding to each of the initial promotional resources, the initial promotional resources are filtered to determine the target promotional resources; wherein, the resource understanding results are obtained in advance by using the generative model to perform content understanding on each of the initial promotional resources; Using the generative model, information is integrated from the target text content and the target promotional resources to generate the target search results.

5. The method according to claim 4, characterized in that, The step of filtering each initial promotional resource based on the target search category, the content understanding result, and the resource understanding result corresponding to each initial promotional resource to determine the target promotional resource includes: Based on the target search category and the resource category in the resource understanding result corresponding to each of the initial candidate resources, category matching is performed to filter out multiple intermediate promotion resources from each of the initial promotion resources; Based on the content understanding results and the remaining understanding results in the resource understanding results corresponding to each intermediate promotion resource, a relevance matching is performed to select at least one candidate promotion resource from each intermediate promotion resource; Based on the content understanding results and the remaining understanding results in the resource understanding results corresponding to each candidate promotion resource, a multi-dimensional sorting and filtering is performed to select at least one target promotion resource from each candidate promotion resource; wherein, the multi-dimensional factors include at least a relevance dimension, a quality dimension, and a resource delivery performance indicator dimension.

6. The method according to claim 5, characterized in that, The process of performing relevance matching based on the content understanding results and the remaining understanding results in the resource understanding results corresponding to each intermediate promotional resource, and selecting at least one candidate promotional resource from each intermediate promotional resource, includes: Based on the content understanding results, user feature data, and the remaining understanding results in the resource understanding results corresponding to each intermediate promotional resource, relevance matching is performed to select at least one candidate promotional resource from each intermediate promotional resource.

7. An information search device, characterized in that, include: The input content determination module is used to determine the input content corresponding to the first interactive operation in response to the information search function. The target search result display module is used to determine the target search result corresponding to the input content and display the input content and the target search result on the search result page. The target search result includes aggregated content matching the input content, and the aggregated content includes the result obtained by aggregating information from the target text content and at least one target promotional resource. The target text content is generated by processing the input content through a generative model. The target promotional resource is obtained by filtering multiple initial promotional resources based on the input content using the generative model. The initial promotional resource is a resource with value conversion attributes, presented as a display-type information carrier, and used to promote a specific target.

8. An electronic device, characterized in that, include: processor; Memory, used to store executable instructions; The processor is configured to read the executable instructions from the memory and execute the executable instructions to implement the information search method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, causes the processor to implement the information search method according to any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product is used to implement the information search method according to any one of claims 1-6.

Images