Information search method and device, storage medium and computer equipment

Abstract

The invention discloses an information search method and device, a storage medium and computer equipment, and relates to the technical field of Internet. The method comprises the steps that a client responds to a received information search request and sends the information search request to a server; the server obtains search content corresponding to the information search request in response to the received information search request; on the basis of the search content, search guide suggestion information is generated through a generative large language model, a search result corresponding to the information search request is obtained, the search guide suggestion information comprises at least one of search description information, guide words and screening items, and the guide words and the screening items are associated with each other; sending the search guide suggestion information and the search result to the client; the client receives and displays the search guide suggestion information and the search result, and the screening item is used for displaying the search result as the search result corresponding to the content in the screening item. According to the method, the information searching efficiency and accuracy can be improved.

Description

Technical Field

[0001] The present invention relates to the field of Internet technologies, and in particular, to an information search method, apparatus, storage medium, and computer device. Background Art

[0002] The information search function is an important function in an e-commerce platform. Through an information search box, a user can input any search content, such as a product name, a product category, a merchant name, etc. Correspondingly, the platform's search engine will provide the user with search results corresponding to the search content for the user to view.

[0003] Currently, as the search content input by users becomes more personalized, much of the search content itself covers very broad information or has very few matching search results. In the prior art, the method of information matching based on keywords in the search content makes it difficult to match search results highly relevant to the above-mentioned search content, resulting in a weak correlation between the search results and the search content and making it difficult to meet the user's search expectations. Summary of the Invention

[0004] In view of this, the present application provides an information search method, apparatus, storage medium, and computer device, mainly aiming to solve the technical problem that it is difficult to match accurate search results for search content covering very broad information or having very few search results, resulting in the search results being difficult to meet the user's search expectations.

[0005] According to a first aspect of the present invention, an information search method is provided, and the method includes:

[0006] The client responds to receiving an information search request and sends the information search request to the server;

[0007] The server responds to receiving the information search request and obtains the search content corresponding to the information search request;

[0008] Based on the search content, the server generates search guidance suggestion information through a generative large language model and obtains search results corresponding to the information search request, where the search guidance suggestion information includes at least one of search description information, guiding words, and filtering options, and the guiding words and the filtering options are mutually related;

[0009] The server sends the search guidance suggestion information and the search results to the client;

[0010] The client receives and displays the search guidance suggestion information and the search results, and the filtering option is used to display the search results as search results corresponding to the content in the filtering option.

[0011] Optionally, the client receives and displays the search guidance suggestion information, including: the client displays at least one of the search description information, the guiding words, and the filtering options in a structured manner, where the search description information and / or the guiding words are displayed in structured text, and the filtering options are displayed through structured controls.

[0012] Optionally, the client displays the filtering options through the structured controls, including: the client displays at least one of the filtering options through controls arranged in sequence, where each control in the sequentially arranged controls is used to display one of the filtering options and is used to, in response to a trigger operation, display the search results as search results corresponding to the content in the triggered filtering option.

[0013] Optionally, the client displays the filtering options through the structured controls, further including: the client displays the text and / or icons corresponding to the filtering options through the controls.

[0014] Optionally, the client receiving and displaying the search guidance suggestion information further includes: the client displays the search description information and / or the guiding words on the first layer of the search result page; and / or, the client displays the filtering options and the search results on the second layer of the search result page.

[0015] Optionally, the search description information includes search suggestions and guiding suggestions, and the guiding suggestions are correlated with the guiding words; the method further includes: the client, in response to an upward sliding operation on the second layer of the search result page, hides some of the guiding suggestions and / or some of the guiding words, and when the second layer is pinned to the top for display, hides all of the guiding suggestions and / or all of the guiding words; and / or the client, in response to a downward sliding operation on the second layer of the search result page, displays the search description information and / or the guiding words, and displays at least one piece of historical search content and the search guidance suggestion information corresponding to the historical search content.

[0016] Optionally, the filtering options are obtained by performing text matching on a preset product name and / or product category and / or merchant category based on the guiding words.

[0017] Optionally, the filtering options are product names and / or product categories and / or merchant categories that match the guiding words; and / or the filtering options are product names and / or product categories and / or merchant categories that match the guiding words, and icons that match the product names and / or product categories and / or merchant categories.

[0018] Optionally, in response to receiving an information search request, the client sends the information search request to the server, including: in response to detecting that the information search request initiated by the user is a preset artificial intelligence search request, the client sends the information search request to the server.

[0019] Optionally, in response to detecting that the information search request initiated by the user is a preset artificial intelligence search request, the client includes: in response to a trigger operation on an artificial intelligence search entry and / or an artificial intelligence entry, the client determines that the information search request initiated by the user is a preset artificial intelligence search request; and / or in the artificial intelligence mode, in response to a sending operation of the search content, the client determines that the information search request initiated by the user is a preset artificial intelligence search request; and / or when the search content input by the user meets a preset condition, in response to a sending operation of the search content, the client determines that the information search request initiated by the user is a preset artificial intelligence search request, where the preset condition includes that a keyword in the search content matches a preset keyword, and / or the number of search results corresponding to the search content is less than a preset number threshold.

[0020] Optionally, the method further includes: in response to receiving pre-search content input by the user, the client sends the pre-search content to the server; the server determines whether a keyword in the pre-search content matches a preset keyword, and / or determines whether the number of search results corresponding to the pre-search content is less than the preset number threshold; when the keyword in the pre-search content matches the preset keyword, and / or the number of search results corresponding to the pre-search content is less than the preset number threshold, the server sends the artificial intelligence search entry and / or the artificial intelligence entry to the client, where the artificial intelligence entry is determined according to the keyword in the pre-search content; the client displays the artificial intelligence search entry and / or the artificial intelligence entry.

[0021] Optionally, the method further includes: in response to a trigger operation on the artificial intelligence search entry and / or the artificial intelligence entry, the client enters the artificial intelligence mode; and / or when the search content input by the user meets the preset condition, in response to at least one of a sending operation, a modification operation, and a replacement operation of the search content, the client enters the artificial intelligence mode.

[0022] Optionally, based on the search content, the server generates search guidance suggestion information through a generative large language model, including: the server inputs the search content, user information initiating the information search request, and a preset prompt template into the generative large language model, so that the generative large language model generates user behavior preference information based on the prompt template and the user information, and generates the search guidance suggestion information based on the prompt template, the user behavior preference information, and the search content.

[0023] Optionally, the prompt template includes a first prompt template and a second prompt template; the server generating search guidance suggestion information through a generative large language model based on the search content further includes: when the number of search results corresponding to the search content is less than a preset quantity threshold, the server generates the search guidance suggestion information through the first prompt template; when the number of search results corresponding to the search content is greater than or equal to the preset quantity threshold, the server generates the search guidance suggestion information through the second prompt template.

[0024] Optionally, the prompt template includes at least one behavior preference dimension and the execution steps corresponding to the behavior preference dimension; the server generating user behavior preference information based on the prompt template and the user information includes: the server generates user behavior preference information for at least one of the behavior preference dimensions based on the user information through the execution steps corresponding to the behavior preference dimension, where the behavior preference dimensions of the user behavior preference information include at least one of keyword preference, product category preference, merchant brand preference, delivery speed preference, order placement time preference, specific product preference, and product attribute preference.

[0025] Optionally, the prompt template further includes a task objective and the execution steps of the task objective; the server generating the search guidance suggestion information based on the prompt template, the user behavior preference information, and the search content includes: the server constructs a semantic extension network based on the keywords in the search content and the related words corresponding to the keywords through the task objective and the execution steps of the task objective, and predicts at least one search extension word and the guidance suggestion corresponding to the search extension word based on the user behavior preference information and / or user location information through the semantic extension network; the server generates the search guidance suggestion information based on the search extension word and the guidance suggestion corresponding to the search extension word.

[0026] Optionally, the prompt template further includes a task output format; the server generates the search guidance suggestion information based on the search expansion term and the guidance suggestion corresponding to the search expansion term, including: the server generates and outputs the search description information, at least one of the guidance terms, and at least one of the screening options according to the task output format, based on the search expansion term and the guidance suggestion corresponding to the search expansion term; and / or the server generates and outputs the search description information and at least one of the guidance terms according to the task output format, based on the search expansion term and the guidance suggestion corresponding to the search expansion term; the server obtains at least one screening option by performing text matching on the guidance term through a preset product name and / or product category and / or merchant category.

[0027] Optionally, the search description information includes a search suggestion and a guidance suggestion, and the guidance suggestion is associated with the guidance term; the client receives and displays the search guidance suggestion information, further including: the client receives the search suggestion and the guidance suggestion in real time and dynamically displays the search suggestion and the guidance suggestion; after the search suggestion and the guidance suggestion are displayed, the client displays each of the guidance terms and the guidance suggestion corresponding to each guidance term in paragraphs with each guidance term as the paragraph start.

[0028] Optionally, the method further includes: after the search guidance suggestion information is displayed, the client controls the device where the client is located to perform a vibration prompt.

[0029] Optionally, the method further includes: before displaying the search guidance suggestion information, the client displays a prompt information indicating that the search guidance suggestion information is being generated, where the prompt information is generated according to a preset prompt information and / or a keyword in the search content; and / or before displaying the search guidance suggestion information, the client dynamically displays an information loading icon; and / or the client displays the search content and an artificial intelligence icon in the information search box, where the artificial intelligence icon is used to identify that the information search request corresponding to the search content is a preset artificial intelligence search request; and / or in response to detecting a network interruption, the client displays an information loading control, where the information loading control is used to re-receive and display the search guidance suggestion information in response to a trigger operation.

[0030] According to a second aspect of the present invention, an information search method is provided, and the method includes:

[0031] In response to receiving an information search request, send the information search request;

[0032] Receive and display search guidance suggestion information and search results, where the search guidance suggestion information is generated by a generative large language model based on the search content, and the search results are obtained through the information search request; the search guidance suggestion information includes at least one of search description information, guiding words, and filtering options; the guiding words and the filtering options are interrelated, and the filtering options are used to display the search results as search results corresponding to the content in the filtering options.

[0033] Optionally, the receiving and displaying the search guidance suggestion information includes: displaying at least one of the search description information, the guiding words, and the filtering options in a structured manner, where the search description information and / or the guiding words are displayed in structured text, and the filtering options are displayed in structured controls.

[0034] Optionally, displaying the filtering options through the structured controls includes: displaying at least one of the filtering options through controls arranged in sequence, where each control in the controls arranged in sequence is used to display one of the filtering options and is used to, in response to a trigger operation, display the search results as search results corresponding to the content in the triggered filtering option.

[0035] Optionally, displaying the filtering options through the structured controls further includes: displaying the text and / or icons corresponding to the filtering options through the controls.

[0036] Optionally, the receiving and displaying the search guidance suggestion information further includes: displaying the search description information and / or the guiding words on the first layer of the search result page; and / or displaying the filtering options and the search results on the second layer of the search result page.

[0037] Optionally, the search description information includes search suggestions and guiding suggestions, and the guiding suggestions are interrelated with the guiding words; the method further includes: in response to an upward sliding operation on the second layer of the search result page, hiding some of the guiding suggestions and / or some of the guiding words, and when the second layer is pinned to the top for display, hiding all of the guiding suggestions and / or all of the guiding words; and / or in response to a downward sliding operation on the second layer of the search result page, displaying the search description information and / or the guiding words, and displaying at least one historical search content and the search guidance suggestion information corresponding to the historical search content.

[0038] Optionally, the filtering options are obtained by performing text matching based on a preset product name and / or product category and / or merchant category based on the guiding words.

[0039] Optionally, the filtering option is a product name and / or product category and / or merchant category that matches the guiding word; and / or the filtering option is a product name and / or product category and / or merchant category that matches the guiding word, and an icon that matches the product name and / or product category and / or merchant category.

[0040] Optionally, the sending of the information search request in response to receiving the information search request includes: in response to detecting that the information search request initiated by the user is a preset artificial intelligence search request, sending the information search request.

[0041] Optionally, the detecting that the information search request initiated by the user is a preset artificial intelligence search request includes: in response to a triggering operation on an artificial intelligence search entry and / or an artificial intelligence entry, determining that the information search request initiated by the user is a preset artificial intelligence search request; and / or in the artificial intelligence mode, in response to the sending operation of the search content, determining that the information search request initiated by the user is a preset artificial intelligence search request; and / or when the search content input by the user meets the preset conditions, in response to the sending operation of the search content, determining that the information search request initiated by the user is a preset artificial intelligence search request, where the preset conditions include that the keywords in the search content match the preset keywords, and / or the number of search results corresponding to the search content is less than a preset quantity threshold.

[0042] Optionally, the method further includes: in response to receiving the pre-search content input by the user, sending the pre-search content; when the keywords in the pre-search content match the preset keywords, and / or the number of search results corresponding to the pre-search content is less than the preset quantity threshold, displaying the artificial intelligence search entry and / or the artificial intelligence entry, where the artificial intelligence entry is determined according to the keywords in the pre-search content.

[0043] Optionally, the method further includes: in response to a triggering operation on the artificial intelligence search entry and / or the artificial intelligence entry, entering the artificial intelligence mode; and / or when the search content input by the user meets the preset conditions, in response to at least one of the sending operation, modification operation, and replacement operation of the search content, entering the artificial intelligence mode.

[0044] Optionally, the search instruction information includes search suggestions and guiding suggestions, and the guiding suggestions are associated with the guiding words; the receiving and displaying the search guiding suggestion information further includes: receiving the search suggestions and the guiding suggestions in real time, and dynamically displaying the search suggestions and the guiding suggestions; after the search suggestions and the guiding suggestions are displayed, using each of the guiding words as the beginning of a paragraph, and displaying each of the guiding words and the guiding suggestions corresponding to the guiding words in paragraphs.

[0045] Optionally, the method further includes: after the search guiding suggestion information is displayed, controlling the device where the client is located to vibrate for prompting.

[0046] Optionally, the method further includes: before displaying the search guiding suggestion information, displaying a prompt message indicating that the search guiding suggestion information is being generated, where the prompt message is generated according to a preset prompt message and / or keywords in the search content; and / or before displaying the search guiding suggestion information, dynamically displaying an information loading icon; and / or displaying the search content and an artificial intelligence icon in the information search box, where the artificial intelligence icon is used to identify that the information search request corresponding to the search content is a preset artificial intelligence search request; and / or in response to detecting a network interruption, displaying an information loading control, where the information loading control is used to, in response to a triggering operation, re-receive and display the search guiding suggestion information.

[0047] According to a third aspect of the present invention, an information search method is provided, and the method includes:

[0048] In response to receiving an information search request, obtaining the search content corresponding to the information request;

[0049] Based on the search content, through a generative large language model, generating search guiding suggestion information, and obtaining a search result corresponding to the information search request, where the search guiding suggestion information includes at least one of search instruction information, guiding words, and screening options, and the guiding words and the screening options are associated with each other;

[0050] Sending the search guiding suggestion information and the search result to the client, and the screening option is used to display the search result as a search result corresponding to the content in the screening option.

[0051] Optionally, the screening option is obtained by text matching based on a preset product name and / or product category and / or merchant category based on the guiding words.

[0052] Optionally, the screening option is a product name and / or product category and / or merchant category that matches the guiding word; and / or the screening option is a product name and / or product category and / or merchant category that matches the guiding word, and an icon that matches the product name and / or product category and / or merchant category.

[0053] Optionally, the method further includes: in response to receiving pre-search content input by a user, determining whether a keyword in the pre-search content matches a preset keyword, and / or determining whether the number of search results corresponding to the pre-search content is less than a preset quantity threshold; when the keyword in the pre-search content matches the preset keyword, and / or the number of search results corresponding to the pre-search content is less than the preset quantity threshold, sending an artificial intelligence search entry and / or an artificial intelligence entry to the client, where the artificial intelligence entry is determined according to the keyword in the pre-search content.

[0054] Optionally, generating search guidance suggestion information based on the search content through a generative large language model includes: inputting the search content, user information for initiating the information search request, and a preset prompt template into the generative large language model, so that the generative large language model generates user behavior preference information based on the prompt template and the user information, and generates the search guidance suggestion information based on the prompt template, the user behavior preference information, and the search content.

[0055] Optionally, the prompt template includes a first prompt template and a second prompt template; generating search guidance suggestion information based on the search content through a generative large language model further includes: when the number of search results corresponding to the search content is less than a preset quantity threshold, generating the search guidance suggestion information through the first prompt template; when the number of search results corresponding to the search content is greater than or equal to the preset quantity threshold, generating the search guidance suggestion information through the second prompt template.

[0056] Optionally, the prompt template includes at least one behavior preference dimension and an execution step corresponding to the behavior preference dimension; generating user behavior preference information based on the prompt template and the user information includes: through the execution step corresponding to the behavior preference dimension, generating user behavior preference information for at least one behavior preference dimension based on the user information, where the behavior preference dimensions of the user behavior preference information include at least one of keyword preference, product category preference, merchant brand preference, delivery speed preference, order placement time preference, specific product preference, and product attribute preference.

[0057] Optionally, the prompt template further includes a task objective and the execution steps of the task objective; generating the search guidance recommendation information based on the prompt template, the user behavior preference information, and the search content includes: based on the task objective and the execution steps of the task objective, constructing a semantic expansion network based on the keywords in the search content and the related words corresponding to the keywords, and predicting at least one search expansion word and the guidance recommendation corresponding to the search expansion word through the semantic expansion network based on the user behavior preference information and / or the user location information; generating the search guidance recommendation information based on the search expansion word and the guidance recommendation corresponding to the search expansion word.

[0058] Optionally, the prompt template further includes a task output format; generating the search guidance recommendation information based on the search expansion word and the guidance recommendation corresponding to the search expansion word includes: generating and outputting the search description information, at least one of the guidance words, and at least one of the filtering options according to the task output format based on the search expansion word and the guidance recommendation corresponding to the search expansion word; and / or generating and outputting the search description information and at least one of the guidance words according to the task output format based on the search expansion word and the guidance recommendation corresponding to the search expansion word; obtaining at least one filtering option by performing text matching on the guidance word through a preset product name and / or product category and / or merchant category.

[0059] According to the fourth aspect of the present invention, an information search device is provided, and the device includes:

[0060] A search request response module, configured to send the information search request in response to receiving an information search request;

[0061] A guidance recommendation display module, configured to receive and display search guidance recommendation information and search results, wherein the search guidance recommendation information is generated by a generative large language model based on the search content, and the search results are obtained through the information search request; the search guidance recommendation information includes at least one of search description information, guidance words, and filtering options; the guidance words and the filtering options are related to each other, and the filtering options are used to display the search results as search results corresponding to the content in the filtering options.

[0062] Optionally, the guidance recommendation display module is specifically configured to: display at least one of the search description information, the guidance words, and the filtering options in a structured manner, wherein the search description information and / or the guidance words are displayed in structured text, and the filtering options are displayed in structured controls.

[0063] Optionally, the guiding suggestion display module is specifically configured to: display at least one of the screening options through sequentially arranged controls, where each of the sequentially arranged controls is used to display one of the screening options and is used to, in response to a triggering operation, display the search results as search results corresponding to the content in the triggered screening option.

[0064] Optionally, the guiding suggestion display module is specifically configured to: display the text and / or icon corresponding to the screening option through the control.

[0065] Optionally, the guiding suggestion display module is specifically configured to: display the search description information and / or the guiding word on the first layer of the search result page; and / or display the screening option and the search results on the second layer of the search result page.

[0066] Optionally, the search description information includes search suggestions and guiding suggestions, and the guiding suggestions are associated with the guiding word; the guiding suggestion display module is further configured to: in response to an upward sliding operation on the second layer of the search result page, hide some of the guiding suggestions and / or some of the guiding words, and when the second layer is pinned to the top for display, hide all of the guiding suggestions and / or all of the guiding words; and / or in response to a downward sliding operation on the second layer of the search result page, display the search description information and / or the guiding word, and display at least one piece of historical search content and the search guiding suggestion information corresponding to the historical search content.

[0067] Optionally, the screening option is obtained by performing text matching on a preset product name and / or product category and / or merchant category based on the guiding word.

[0068] Optionally, the screening option is a product name and / or product category and / or merchant category that matches the guiding word; and / or the screening option is a product name and / or product category and / or merchant category that matches the guiding word, and an icon that matches the product name and / or product category and / or merchant category.

[0069] Optionally, the search request response module is specifically configured to: in response to detecting that an information search request initiated by a user is a preset artificial intelligence search request, send the information search request.

[0070] Optionally, the search request response module is specifically configured to: in response to a triggering operation on an artificial intelligence search entry and / or an artificial intelligence entry, determine that an information search request initiated by a user is a preset artificial intelligence search request; and / or in an artificial intelligence mode, in response to a sending operation of the search content, determine that the information search request initiated by the user is a preset artificial intelligence search request; and / or when the search content input by the user meets a preset condition, in response to the sending operation of the search content, determine that the information search request initiated by the user is a preset artificial intelligence search request, where the preset condition includes that a keyword in the search content matches a preset keyword, and / or the number of search results corresponding to the search content is less than a preset quantity threshold.

[0071] Optionally, the search request response module is further configured to: in response to receiving pre-search content input by the user, send the pre-search content; when a keyword in the pre-search content matches the preset keyword, and / or the number of search results corresponding to the pre-search content is less than the preset quantity threshold, display the artificial intelligence search entry and / or the artificial intelligence entry, where the artificial intelligence entry is determined according to the keyword in the pre-search content.

[0072] Optionally, the search request response module is further configured to: in response to a triggering operation on the artificial intelligence search entry and / or the artificial intelligence entry, enter the artificial intelligence mode; and / or when the search content input by the user meets the preset condition, in response to at least one of a sending operation, a modification operation, and a replacement operation of the search content, enter the artificial intelligence mode.

[0073] Optionally, the search description information includes search suggestions and guiding suggestions, and the guiding suggestions are associated with the guiding words; the guiding suggestion display module is specifically further configured to: receive the search suggestions and the guiding suggestions in real time and dynamically display the search suggestions and the guiding suggestions; after the search suggestions and the guiding suggestions are displayed, use each of the guiding words as the beginning of a paragraph to display each of the guiding words and the guiding suggestions corresponding to the guiding words in a segmented manner.

[0074] Optionally, the guiding suggestion display module is further configured to: after the search guiding suggestion information is displayed, control the device where the client is located to perform a vibration prompt.

[0075] Optionally, the search guidance display module is further configured to: before displaying the search guidance suggestion information, display a prompt message indicating that the search guidance suggestion information is being generated, where the prompt message is generated according to a preset prompt message and / or keywords in the search content; and / or before displaying the search guidance suggestion information, dynamically display an information loading icon; and / or display the search content and an artificial intelligence icon in the information search box, where the artificial intelligence icon is used to identify that the information search request corresponding to the search content is a preset artificial intelligence search request; and / or in response to detecting a network interruption, display an information loading control, where the information loading control is used to, in response to a trigger operation, re-receive and display the search guidance suggestion information.

[0076] According to a fifth aspect of the present invention, there is provided an information search device, which includes:

[0077] A search request response module, configured to, in response to receiving an information search request, obtain the search content corresponding to the information request;

[0078] A guidance suggestion generation module, configured to, based on the search content, generate search guidance suggestion information through a generative large language model, and obtain a search result corresponding to the information search request, where the search guidance suggestion information includes at least one of search description information, guiding words, and filtering options, and the guiding words and the filtering options are interrelated;

[0079] A guidance suggestion sending module, configured to send the search guidance suggestion information and the search result to the client, and the filtering option is used to display the search result as a search result corresponding to the content in the filtering option.

[0080] Optionally, the filtering option is obtained by performing text matching on a preset product name and / or product category and / or merchant category based on the guiding words.

[0081] Optionally, the filtering option is a product name and / or product category and / or merchant category that matches the guiding words; and / or the filtering option is a product name and / or product category and / or merchant category that matches the guiding words, and an icon that matches the product name and / or product category and / or merchant category.

[0082] Optionally, the search request response module is further configured to: in response to receiving the pre-search content input by the user, determine whether the keywords in the pre-search content match the preset keywords, and / or determine whether the number of search results corresponding to the pre-search content is less than the preset quantity threshold; when the keywords in the pre-search content match the preset keywords, and / or the number of search results corresponding to the pre-search content is less than the preset quantity threshold, send the artificial intelligence search entry and / or the artificial intelligence entry to the client, where the artificial intelligence entry is determined according to the keywords in the pre-search content.

[0083] Optionally, the guidance and suggestion generation module is specifically configured to: input the search content, the user information for initiating the information search request, and a preset prompt template into the generative large language model, so that the generative large language model generates user behavior preference information based on the prompt template and the user information, and generates the search guidance and suggestion information based on the prompt template, the user behavior preference information, and the search content.

[0084] Optionally, the prompt template includes a first prompt template and a second prompt template; the guidance and suggestion generation module is further specifically configured to: generate the search guidance and suggestion information through the first prompt template when the number of search results corresponding to the search content is less than the preset quantity threshold; generate the search guidance and suggestion information through the second prompt template when the number of search results corresponding to the search content is greater than or equal to the preset quantity threshold.

[0085] Optionally, the prompt template includes at least one behavior preference dimension and the execution steps corresponding to the behavior preference dimension; the guidance and suggestion generation module is specifically configured to: generate user behavior preference information of at least one of the behavior preference dimensions based on the user information through the execution steps corresponding to the behavior preference dimension, where the behavior preference dimensions of the user behavior preference information include at least one of keyword preference, product category preference, merchant brand preference, delivery speed preference, order placement time preference, specific product preference, and product attribute preference.

[0086] Optionally, the prompt template further includes a task objective and the execution steps of the task objective; the guidance and suggestion generation module is specifically configured to: construct a semantic expansion network based on the keywords in the search content and the related words corresponding to the keywords through the task objective and the execution steps of the task objective, and predict at least one search expansion word and the guidance suggestion corresponding to the search expansion word through the semantic expansion network based on the user behavior preference information and / or the user location information; generate the search guidance and suggestion information based on the search expansion word and the guidance suggestion corresponding to the search expansion word.

[0087] Optionally, the guidance suggestion generation module is specifically configured to: generate and output the search description information, at least one of the guidance words, and at least one of the screening options according to the task output format, based on the search expansion word and the guidance suggestion corresponding to the search expansion word; and / or generate and output the search description information and at least one of the guidance words according to the task output format, based on the search expansion word and the guidance suggestion corresponding to the search expansion word; and obtain at least one screening option by performing text matching on the guidance word through a preset product name and / or product category and / or merchant category.

[0088] According to a sixth aspect of the present invention, there is provided a storage medium having a computer program stored thereon, and when the program is executed by a processor, the above information search method is implemented.

[0089] According to a seventh aspect of the present invention, there is provided a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, the above information search method is implemented.

[0090] An information search method, device, storage medium, and computer device provided by the present invention first respond to an information search request initiated by a user, send the information search request to a server so that the server obtains search content corresponding to the information search request, and then, according to the search content, generate, through a generative large language model, search guidance suggestion information covering various information such as search description information, guidance words, and screening options, and obtain search results corresponding to the information search request. Finally, send the generated search guidance suggestion information and the obtained search results to the client for display. By combining the generative large language model with the information search scenario, the above method can decompose and extract information from search content with relatively broad coverage or very few retrieval results through the text analysis and text generation capabilities of the generative large language model, so as to generate search description information, guidance words, screening options, etc. that can guide users to conduct further information searches. By displaying the search guidance suggestion information, the client can provide more accurate and personalized search results for users, and present the search results to users in a more interpretable manner, thereby improving the accuracy of the search results and enhancing the user's information search experience. In addition, the above information search method does not interfere with the normal information search link, and users can still view search results corresponding to the information search request by initiating the information search request.

[0091] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the following specific embodiments of this application are specifically given. Brief Description of the Drawings

[0092] The drawings described herein are used to provide a further understanding of the present invention, and constitute a part of this application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings:

[0093] Figure 1 A flowchart showing a method for information search provided by an embodiment of the present invention is shown;

[0094] Figure 2 A schematic diagram of a scenario of a method for information search provided by an embodiment of the present invention is shown;

[0095] Figure 3 A schematic diagram of a scenario of a method for information search provided by an embodiment of the present invention is shown;

[0096] Figure 4 A schematic diagram of a scenario of a method for information search provided by an embodiment of the present invention is shown;

[0097] Figure 5 A schematic diagram of a scenario of a method for information search provided by an embodiment of the present invention is shown;

[0098] Figure 6 A flowchart showing a method for information search provided by an embodiment of the present invention is shown;

[0099] Figure 7 A flowchart showing a method for information search provided by an embodiment of the present invention is shown;

[0100] Figure 8 A schematic diagram of the structure of an information search device provided by an embodiment of the present invention is shown;

[0101] Figure 9 A schematic diagram of the structure of an information search device provided by an embodiment of the present invention is shown;

[0102] Figure 10 A schematic diagram of the structure of a computer device provided by an embodiment of the present invention is shown. Detailed Embodiments

[0103] The present invention will be described in detail below with reference to the drawings and in conjunction with the embodiments. It should be noted that, without conflict, the embodiments in this application and the features in the embodiments can be combined with each other.

[0104] In one embodiment, as Figure 1 shown, an information search method is provided. Taking the case where this method is applied to a client and a server as an example, it includes the following steps:

[0105] 101. In response to receiving an information search request, the client sends the information search request to the server.

[0106] Specifically, when the user enters a certain search content or selects a certain search content and sends it through the information search box of the client, the client can detect that the user has initiated an information search request. After detecting that the user has initiated an information search request, the client can send the information search request to the server.

[0107] In this embodiment, the user can initiate an information search request in various ways. For example, the user can initiate an information search request by selecting a certain search content, or by inputting and sending a paragraph of search content, or by clicking a pre-set search entry. In addition, the information search request initiated by the user can be a pre-set artificial intelligence search request or an ordinary information search request. Regardless of which type of information search request the user initiates, the information search request will be sent to the server.

[0108] 102. In response to receiving the information search request, the server obtains the search content corresponding to the information search request.

[0109] Among them, the search content refers to the content input or selected by the user when initiating an information search request. This content can be search keywords such as product names, product categories, merchant names, or phrases arbitrarily input by the user, such as "what is suitable to eat for a cold" and "what to eat for dinner". Generally speaking, for search content with clear search keywords, the search engine will recall relatively corresponding search results more accurately for the user. At this time, the correlation between the search results and the search content is relatively high, and the search accuracy is relatively high. In contrast, for phrases arbitrarily input by the user, the search engine usually cannot accurately match, the correlation between the search results and the search content is relatively low, and the search accuracy rate is also relatively low. In addition, for some infrequently used search keywords, such as "a certain brand of milk tea", since there is no such brand of milk tea within the delivery range of the user's location, the number of retrieved results is also very small or zero (few or zero results). At this time, the search results recalled by the search engine also have a relatively low correlation with the search content, and the search accuracy rate is also relatively low.

[0110] Specifically, in response to receiving the information search request, the server can obtain the search content corresponding to the information search request. Among them, the search content can be carried and sent in the information search request, or can be sent to the server separately. This embodiment does not make specific limitations on the acquisition method of the search content.

[0111] 103. The server generates search guidance suggestion information through a generative large language model based on the search content and obtains search results corresponding to the information search request.

[0112] 104. The server sends the search guidance suggestion information and the search results to the client.

[0113] Among them, the search guidance suggestion information includes at least one of search description information, guiding words, and filtering options. Among them, the search description information refers to guiding explanation information of the suggested reasons for the search content and / or extended keywords related to the search content with a certain word limit. For example, assuming the search content is "what is suitable to eat for a cold", the search description information can be "it is suitable to eat some light foods for a cold", "chicken soup is rich in protein and amino acids, which can improve immunity and anti-inflammatory effects"; assuming the search content is "what to have for breakfast", the search description information can be "you can have some nutritious foods for breakfast", "baozi (steamed buns) and porridge can be an ideal choice for a nutritious breakfast"; assuming the search content is "XX milk tea" and the current search results are few, the search description information can be "the current search results are few, looking for some inspiration for drinking milk tea for you", "Hong Kong-style milk tea, as a classic milk tea category, might be worth a try".

[0114] Furthermore, the guiding words refer to extended keywords generated based on the search content. The extended keywords have a certain relevance to the search content and the search description information and can be directly used as search keywords to guide users to conduct further information searches, generally can be product names, product categories, merchant categories, etc. For example, assuming the search content is "what is suitable to eat for a cold", the search description information is "it is suitable to eat some light foods for a cold", "chicken soup is rich in protein and amino acids, which can improve immunity and anti-inflammatory effects; green vegetables help enhance immunity", then the guiding words can be light product categories or product names such as "chicken soup, green vegetables". Through the guiding words, users can initiate an information search request with a more specific search intention to obtain accurate search results.

[0115] Furthermore, the screening option refers to an extended keyword associated with the guiding word. This screening option can, in response to a user's triggering operation, initiate an information search request directly using the extended keyword corresponding to the screening option as the search keyword, thereby obtaining search results corresponding to the extended keyword. Among them, the screening word can be a product name, a product category, a merchant category, etc. For example, assuming the search content is "what is suitable to eat for a cold", the search description information is "it is suitable to eat some light foods for a cold", "chicken soup is rich in protein and amino acids, which can improve immunity and have an anti-inflammatory effect; green vegetables help enhance immunity", and the guiding words are product category names such as "chicken soup, green vegetables, tomatoes", then the screening options can be product names such as "old hen soup, boiled green vegetables, scrambled eggs with tomatoes". By clicking on the screening option "scrambled eggs with tomatoes", the user can directly view the search results corresponding to "scrambled eggs with tomatoes", thereby improving the information search efficiency.

[0116] Specifically, after the server obtains the search content corresponding to the information search request, it can preprocess the search content and input the search content into the generative large language model in a certain way, so as to perform related processing such as keyword expansion, keyword prediction, and guiding suggestion generation based on the search content through the generative large language model, thereby obtaining search guiding suggestion information that can guide the user to conduct further information search, and then sending the generated search guiding suggestion information to the client.

[0117] In this embodiment, in the process of inputting the search content into the generative large language model, a pre-set prompt template can be borrowed for information input, so that the generative large language model can better understand the task objective and confirm the steps of task execution, thereby being able to output search guiding suggestion information that is more helpful for the user to conduct information search. In addition, since the search content is input by the user according to their own behavior habits and its content has a high degree of randomness, therefore, for different scenarios, prompt templates for multiple scenarios can be designed to improve the accuracy of information output.

[0118] In addition, after the server receives the information search request, it can also obtain the search results corresponding to the information search request. Among them, obtaining the search results corresponding to the information search request can be directly obtaining search results based on the search content; it can also be obtaining search results according to some information in the search guiding suggestion information generated from the search content; it can also be obtaining search results according to the search content and / or any intermediate or result information generated from the search content. It can be understood that, in the case of no conflict and no mutual dependence, the order of generating search guiding suggestion information and obtaining search results is not restricted; in the case of dependence, they can also be executed successively. Furthermore, after obtaining the search results corresponding to the information search request, the server can send the search results to the client.

[0119] 105. The client receives and displays search guidance suggestion information and search results. The filtering option is used to display the search results as search results corresponding to the content in the filtering option.

[0120] Specifically, after the client receives the search guidance suggestion information and search results sent by the server, it can display the search guidance suggestion information and search results on a preset page of the client. Among them, the preset page can be various pages such as a human-computer dialogue page, a search result page, etc. The search description information, guiding words, filtering options, and search results in the search guidance suggestion can be displayed at different positions on the preset page in a preset manner. In this embodiment, the filtering option in the search guidance suggestion can be used to update and display the search results displayed on the current page as search results corresponding to the content in the filtering option.

[0121] In this embodiment, the client can receive the search guidance suggestion information in real time and dynamically display the received search guidance suggestion information in real time. It can also display all the search guidance suggestion information completely according to the preset layout after receiving the search guidance suggestion information. In addition, the client can receive and display some of the information in the search guidance suggestion information, or receive and display all the search guidance suggestion information. In some scenarios, the search guidance suggestion information displayed by the client may be blocked or hidden under certain operations of the user. To ensure that the user can see the search guidance suggestion information, the client can display at least part of the search guidance suggestion information. For example, the client can only display the search suggestions and filtering options and hide the guiding words, so as to improve the efficiency of the user to obtain the search guidance suggestion information.

[0122] The information search method provided in this embodiment first responds to an information search request initiated by a user, sends the information search request to a server so that the server can obtain the search content corresponding to the information search request, then generates search guidance suggestion information covering various information such as search description information, guiding words, and filtering options through a generative large language model according to the search content, and obtains search results corresponding to the information search request. Finally, the generated search guidance suggestion information and the obtained search results are sent to the client for display. By combining the generative large language model with the information search scenario, the above method can decompose and extract information from search content with relatively broad coverage or very few retrieval results through the text analysis and text generation capabilities of the generative large language model, so as to generate search description information, guiding words, filtering options, etc. that can guide users to conduct further information searches. By displaying the search guidance suggestion information, the client can provide more accurate and personalized search results for users and present the search results to users in a more interpretable manner, thereby improving the accuracy of the search results and enhancing the user's information search experience. In addition, the above information search method does not interfere with the normal information search link, and users can still view the search results corresponding to the information search request by initiating the information search request.

[0123] In one embodiment, step 101 can be implemented in the following way: The client responds to detecting that the information search request initiated by the user is a preset artificial intelligence search request and sends the information search request to the server.

[0124] Among them, the preset artificial intelligence search request refers to an information search request that can trigger the server to start the generative large language model to conduct search guidance on the search content input by the user. Except for the preset artificial intelligence search request, other information search requests are information search requests that execute the ordinary search link. After the information search request that executes the ordinary search link is sent to the server, the server can obtain the search results corresponding to the information search request. In the process of obtaining the search results, relevant means of artificial intelligence may be borrowed. However, this request that borrows the means of artificial intelligence to obtain search results is not the preset artificial intelligence search request referred to in this embodiment, but other types of artificial intelligence search requests.

[0125] Specifically, when the user enters a certain search content or selects a certain search content to send through the information search box of the client, the client can detect that the user has initiated an information search request. After detecting that the user has initiated an information search request, the client can determine whether the information search request initiated by the user is a preset artificial intelligence search request. If it is detected that the information search request initiated by the user is a preset artificial intelligence search request, the client can send the information search request to the server.

[0126] In this embodiment, the information search request initiated by the user can be detected as a preset artificial intelligence search request in various ways. Among them, the preset artificial intelligence search request can be divided into two initiation methods: active and passive. For example, the client can actively initiate a preset artificial intelligence search request by triggering the artificial intelligence search entry and / or artificial intelligence entry displayed on the client, or passively determine the information search request initiated by the user as a preset artificial intelligence search request when it detects that the search content input by the user meets the initiation conditions of the preset artificial intelligence search request. Correspondingly, the client can also detect whether the information search request initiated by the user is a preset artificial intelligence search request in various ways. For example, the client can identify it through the initiation path of the information search request, the artificial intelligence identifier carried in the information search request, etc., and no specific limitation is made here.

[0127] In this embodiment, by sending the information search request to the server when it detects that the information search request initiated by the user is a preset artificial intelligence search request, the diversion speed of the information search request can be improved. Specifically, when it detects that the information search request is not a preset artificial intelligence search request, the information search request can be diverted to the search engine in the server to obtain the search results corresponding to the information search request; when it detects that the information search request is a preset artificial intelligence search request, the information search request is sent to the search engine in the server and the interface of the generative large language model respectively to obtain the search results corresponding to the information search request and the search guidance suggestion information generated based on the search content respectively.

[0128] In one embodiment, the method of detecting whether the information search request initiated by the user in step 101 is a preset artificial intelligence search request can be implemented by at least one of the following methods:

[0129] 1011. The client determines that the information search request initiated by the user is a preset artificial intelligence search request in response to the trigger operation of the artificial intelligence search entry and / or artificial intelligence entry.

[0130] Specifically, the method of step 1011 is a method of actively initiating a preset artificial intelligence search request. Among them, the artificial intelligence search entry can be designed in a fixed style or in a style combined with the artificial intelligence entry, and the artificial intelligence entry can be obtained based on the search content input by the user. Among them, the artificial intelligence entry can be obtained by text matching based on the keywords in the search content input by the user, or directly generated by the generative large language model based on the search content input by the user. It can be understood that the setting method and setting position of the artificial intelligence search entry and / or artificial intelligence entry can be set according to the actual situation, and no specific limitation is made in this embodiment.

[0131] For example,Figure 3 As shown in the figure on the left, when the user clicks on the artificial intelligence search entry "View Recommendations", or clicks on the artificial intelligence entry "What foods should I eat for a cold?", or clicks on the artificial intelligence search entry "What foods should I eat for a cold? View Recommendations" to initiate an information search operation, the client can, in response to the trigger operation of the artificial intelligence search entry or the artificial intelligence entry, determine that the information search request initiated by the user is a preset artificial intelligence search request, and then send the information search request to the server.

[0132] 1012. In the artificial intelligence mode, in response to the sending operation of the search content, the client determines that the information search request initiated by the user is a preset artificial intelligence search request.

[0133] Specifically, the method of step 1012 is a method of actively initiating a preset artificial intelligence search request, and the user can enter the artificial intelligence mode in an active or passive manner. For example, the user can actively enter the artificial intelligence mode by clicking on the artificial intelligence search entry and / or the artificial intelligence entry, or can passively enter the artificial intelligence mode when the input search content meets the preset conditions. In the artificial intelligence mode, when the user enters any keyword or phrase and initiates an information search request, the preset artificial intelligence search request can be triggered again. It can be understood that the methods of entering or exiting the artificial intelligence mode can be set according to the actual situation, and are not specifically limited in this embodiment.

[0134] For example, as Figure 3 shown, when the user clicks on the artificial intelligence search entry "What foods should I eat for a cold? View Recommendations" to initiate an information search operation, the client can enter the artificial intelligence mode. In the artificial intelligence mode, if the user changes the search content "What foods should I eat for a cold?" to "What should I have for breakfast" and initiates an information search request, the client can still determine that the information search request currently initiated by the user is a preset artificial intelligence search request, and then send the information search request to the server.

[0135] 1013. When the search content input by the user meets the preset conditions, in response to the sending operation of the search content, the client determines that the information search request initiated by the user is a preset artificial intelligence search request.

[0136] Specifically, the method in step 1013 is a method of passively initiating a preset artificial intelligence search request. Among them, the preset conditions may include at least one of the following two conditions: the keyword in the search content matches the preset keyword, and the number of search results corresponding to the search content is less than the preset quantity threshold. Under the first condition, if the keyword in the search content matches the preset keyword, it can be determined that the keyword in the search content conforms to the characteristics of the preset broad keyword, such as "cold, breakfast, lunch, dinner, sour and sweet, sweet and spicy", etc. At this time, the relevance between the search result and the search content is relatively low. Therefore, the search content can be guided by artificial intelligence to improve the accuracy of information search. Among them, the preset keyword can be pre-stored in the artificial intelligence offline word list for easy acquisition and matching. Further, under the second condition, if the number of search results corresponding to the search content is less than the preset quantity threshold, it can be determined that the number of search results corresponding to the current search content is small, and the relevance between the search result and the search content is also relatively low. At this time, the search content can also be guided by artificial intelligence to improve the accuracy of information search. It can be understood that the process of condition judgment can be executed on the client side or on the server side. In addition, the preset conditions can also be set as other conditions, which are not specifically limited in this embodiment.

[0137] For example, as Figure 2 shown, after the user enters the search content "XX milk tea", the client detects that there is only one search result for this brand of milk tea, and this quantity is less than the preset quantity threshold. Then, after the information search request is sent, the client can determine that the information search request initiated by the user is a preset artificial intelligence search request based on this condition. For another example, after the user enters the search content "What foods should I eat when having a cold?", the client detects that this search content matches the preset keywords "cold, food". Then, after the information search request is sent, the client can also determine that the information search request initiated by the user is a preset artificial intelligence search request based on this condition, and then send this information search request to the server.

[0138] The above embodiments can detect whether the information search request initiated by the user is a preset artificial intelligence search request in various ways, and can enable the user to participate in the artificial intelligence search scenario in various ways, thereby improving the user's perception and participation rate of artificial intelligence search. And, this embodiment can improve the search accuracy by initiating a preset artificial intelligence search request, improve the problem of low hit rate of traditional search results, and thus improve the user's search experience.

[0139] In one embodiment, before step 101, the above information search method further includes the following steps:

[0140] 201. In response to receiving the pre-search content input by the user, the client sends the pre-search content to the server.

[0141] The client can send the pre-search content to the server while the user is inputting the pre-search content in real time. For example, when the user inputs the word "感", the client can send the word "感" to the server, and when the user further inputs the word "冒", the client can send "寒寒" to the server.

[0142] 202. The server determines whether the keyword in the pre-search content matches the preset keyword, and / or determines whether the number of search results corresponding to the pre-search content is less than a preset number threshold.

[0143] The way in which the server performs conditional judgment can be found in step 1013 and will not be described in detail here.

[0144] 203. When the keywords in the pre-search content match the preset keywords, and / or the number of search results corresponding to the search content is less than a preset number threshold, the server sends the artificial intelligence search entry and / or artificial intelligence terms to the client.

[0145] Among them, the artificial intelligence entry can be determined based on the keywords in the pre-search content. In this embodiment, the artificial intelligence entry can be obtained by text matching based on the pre-search content input by the user, or it can be directly generated by a generative large language model based on the pre-search content input by the user. For example, when the pre-search content input by the user is "cold", the artificial intelligence entry can be "what to eat for a cold", or "what medicine to take for a cold", etc.

[0146] In this embodiment, when the server detects that the keywords in the pre-search content match the preset keywords, and / or the number of search results corresponding to the pre-search content is less than a preset number threshold, it can also send an artificial intelligence identifier to the client. Based on the artificial intelligence identifier, the client can determine that the current information search request is a preset artificial intelligence search request, thereby improving the detection speed of the preset artificial intelligence search request.

[0147] 204. The client displays an artificial intelligence search portal and / or artificial intelligence terms.

[0148] Among them, refer to Figure 3For the picture on the left, the client can display an artificial intelligence search entry and / or artificial intelligence entries in the search association page, so that users can initiate a preset artificial intelligence search request during the search process. The client can also display an artificial intelligence search entry and / or artificial intelligence entries in other pages. For example, the client can display an artificial intelligence search entry at a fixed position on the home page, so that users can directly initiate a preset artificial intelligence search request on the home page. It can be understood that the display method and display position of the artificial intelligence search entry and / or artificial intelligence entries are not specifically limited in this embodiment.

[0149] In the above embodiment, by judging the pre-search content in real time during the process of the user inputting the pre-search content and displaying the artificial intelligence entry and / or artificial intelligence entries when the preset conditions are met, the user can view the initiation entry of the preset artificial intelligence search request or the relevant entries for initiating the preset artificial intelligence search request in a timely manner during the process of initiating an information search request, thereby improving the perception and participation rate of artificial intelligence search. Moreover, by initiating the preset artificial intelligence search request, the search accuracy can be improved, and the problem of low hit rate of traditional search results can be solved, so as to improve the user's search experience.

[0150] In one embodiment, the information search method may further include at least one of the following methods:

[0151] 205. The client enters the artificial intelligence mode in response to the triggering operation of the artificial intelligence search entry and / or artificial intelligence entries.

[0152] 206. When the search content input by the user meets the preset conditions, the client enters the artificial intelligence mode in response to at least one of the operations of sending, modifying, and replacing the search content.

[0153] Specifically, when the user clicks on the artificial intelligence search entry and / or artificial intelligence entry displayed on the client, or when the input search content meets the preset conditions, the client can control the application to enter the artificial intelligence mode. In the artificial intelligence mode, the user can freely change and replace the search content. Moreover, after the user sends the changed and replaced search content, the client can detect that the information search request initiated by the user is a preset artificial intelligence search request, and then send this information search request to the server. It can be understood that in other embodiments, when the user clicks on the artificial intelligence search entry and / or artificial intelligence entry, or when the input search content meets the preset conditions, the client can also only determine the information search request currently initiated by the user as a preset artificial intelligence search request without entering the artificial intelligence mode. Then, when the user clicks on the artificial intelligence search entry and / or artificial intelligence entry again next time, or when the input search content meets the preset conditions, the information search request initiated next time is determined as a preset artificial intelligence search request.

[0154] In the above embodiment, when the user actively or passively initiates a preset artificial intelligence search request, it means that the user may not have a clear search intention at this time. In this scenario, the user may continuously change their search content until they view the search results they are interested in. For this scenario, in this embodiment, by directly entering the artificial intelligence mode after detecting that the user has initiated a preset artificial intelligence search request, the user can freely change or replace their search content in the artificial intelligence mode. Moreover, for each input search content, the client can determine that the information search request initiated by the user is a preset artificial intelligence search request, thereby guiding the user to conduct a more targeted information search, improving the user's search efficiency, and enhancing the user's search experience.

[0155] In one embodiment, the method by which the server generates search guidance suggestion information in step 103 can be implemented by the following method:

[0156] 1031. The server inputs the search content, the user information for initiating the information search request, and a preset prompt template into the generative large language model, so that the generative large language model generates user behavior preference information based on the prompt template and user information.

[0157] 1032. The generative large language model running on the server generates search guidance suggestion information based on the prompt template, user behavior preference information, and search content.

[0158] Before the server generates search guidance suggestion information by searching the content, it may first obtain a prompt template for generating search guidance suggestion information, i.e., prompt, and obtain user information of the user who initiated the information search request. The user information may be obtained from a database or server log with the authorization of the user. In this embodiment, the user information may include at least one of user behavior information, user portrait information, user basic information, and the like.

[0159] Specifically, in the process of generating search guidance suggestion information through a generative large language model (hereinafter referred to as the large model), user behavior preference information can be generated with the help of a preset prompt template, such as generating user preferences for keywords, product names, product categories, and merchant brands. Then, the user preference information is combined with the search content input by the user to predict the search results that the user is interested in, so as to obtain search guidance suggestion information that is closer to the search content and more relevant to the user's interests.

[0160] The above-mentioned embodiment generates user behavior preference information with the help of preset prompt templates and user information, and then generates search guidance suggestion information through user behavior preference information and search content. This can make it easier for the big model to understand user behavior preferences, task objectives, and the execution steps of task objectives, thereby being able to generate search guidance suggestion information that is closer to the search content and closer to user preferences, thereby improving the accuracy of the search guidance suggestion information and its matching degree with users, and improving the user search experience.

[0161] In one embodiment, the prompt template may include at least one behavior preference dimension and execution steps corresponding to the behavior preference dimension. Then step 1031 may be implemented by the following method: the server generates user behavior preference information of at least one behavior preference dimension based on user information through the execution steps corresponding to the behavior preference dimension, wherein the behavior preference dimension of the user behavior preference information may include at least one of keyword preference, product category preference, merchant brand preference, delivery speed preference, order time preference, specific product preference and product attribute preference.

[0162] Specifically, the prompt template may include at least one behavior preference dimension and the execution steps corresponding to the behavior preference dimension, so that the large model can quantitatively analyze the user information through multiple behavior preference dimensions and the execution steps corresponding to each behavior preference dimension, so as to obtain user behavior preference information that matches the search scenario. In addition, the prompt template may also include role information so that the model can understand the task goal more accurately. For example, the prompt template may include the following content:

[0163] 1. Role Setting

[0164] 1) Role: You are an expert in food ordering, proficient in providing guidance and suggestions according to different broad intention scenarios, but note that you will not give advice related to medications and dietary therapies.

[0165] 2) Personality: You present an image of an expert with plain language and no use of professional jargon.

[0166] 2. Task Training

[0167] Based on the user's food ordering records on the food delivery platform in the recent 7 days / 14 days / 30 days, summarize the user's food ordering preferences during this period according to the given format, and put forward corresponding suggestions based on their eating patterns.

[0168] 1) Behavior Summary

[0169] User search preferences: In the past 7 days / 14 days / 30 days, the user's preferences for using the search engine include, but are not limited to, the following dimensions:

[0170] 1. Keyword preferences: The user may tend to use specific keywords or phrases when expressing search intentions.

[0171] 2. Food category preferences: Statistically analyze the food category with the highest ordering frequency of the user, such as Chinese cuisine, Western cuisine, fast food, healthy light meals, etc.

[0172] 3. Merchant brand preferences: Collect the merchant brands with the most orders placed by the user to identify brand loyalty.

[0173] 4. Logistics delivery speed: Calculate the average delivery time of the user's orders and analyze the user's sensitivity to delivery speed.

[0174] 5. Regularity of dining time: Organize the user's food ordering times and analyze their daily eating patterns, such as breakfast time period, lunch time period, and dinner time period.

[0175] 6. Preferences for specific dishes: List the specific dishes that the user often orders to understand their preferences for specific foods.

[0176] 7. Dietary taste preferences: Extract the taste types preferred by the user from the food ordering options, such as spicy, sweet, light, or preferences for various special seasonings. ......

[0178] In the above embodiments, by setting the execution steps corresponding to the behavior preference dimensions in the prompt template, it is convenient for the large model to accurately generate user behavior preference information in multiple dimensions that matches the user information through the execution steps corresponding to each preference dimension. Moreover, by setting the preference dimensions in advance, it is possible to prevent the large model from having hallucinations or generating unstable content, thereby making the user behavior preference information more matching the search scenario. At the same time, it is also convenient to obtain search guidance suggestion information that matches the user's own characteristics based on the user behavior preference information, thereby improving the accuracy and matching degree of information search.

[0179] In one embodiment, the prompt template may further include a task objective and the execution steps of the task objective. Then, step 1032 can be implemented by the following method: The server constructs a semantic expansion network based on the keywords and related words corresponding to the keywords in the search content through the task objective and the execution steps of the task objective, and predicts at least one search expansion word and the guidance suggestion corresponding to the search expansion word through the semantic expansion network based on the user behavior preference information and / or the user location information. Based on the search expansion word and the guidance suggestion corresponding to the search expansion word, search guidance suggestion information is generated.

[0180] Specifically, the prompt template may further include a task objective and the execution steps of the task objective, so that the large model can accurately obtain search guidance suggestion information that matches the task objective through the execution steps of the task objective based on the user behavior preference information and the search content input by the user. For example, continuing with the example of the previous prompt template, the prompt template may further include the following content: ......

[0182] 2) Objective: Give keyword guidance suggestions

[0183] Use synonyms, hyponyms, related words, etc. of the user input phrase to construct a semantic expansion network. And predict the search expansion words that the user may want through the context. Combine the collection of the user's context information, such as geographical location, search history preferences, and give guidance on the keywords:

[0184] Guidance suggestions for broad search of dining words:

[0185] 1) Key phrase: Reason for the suggestion, briefly describe the reason in one sentence

[0186] 2) Guiding words: Put forward at least three guiding suggestions based on the user's search key related words ......

[0188] In the above embodiments, the related words corresponding to the keywords in the search content can be various words such as synonyms, hypernyms, hyponyms, and related words of the keywords. For example, assuming the input content is "what foods to eat for a cold", the keywords in the search content can include "cold" and "foods", and based on this, various words such as "being sick", "having a fever", "diet", and "dishes" can be extended. Through the extended words, a semantic expansion network can be constructed. Based on this semantic expansion network, combined with the user's preferences for foods, brands, and keywords in the past period of time, and combined with the user's location, search expansion words such as "chicken soup, green vegetables, tomatoes" and the guiding suggestions corresponding to the search expansion words can be predicted, so as to obtain search guiding suggestion information that matches the search content and the user's behavior preferences.

[0189] In the above embodiments, by constructing a semantic expansion network with the related words corresponding to the keywords in the search content, and on the basis of the semantic expansion network, combining the user's preferences for foods, brands, and keywords in the past period of time, and then combining the user's location, the search expansion words can be predicted, so that the predicted keywords are no longer limited to the search content itself, but can be extended to the entire semantic network, thereby effectively expanding the scope of keyword prediction. In addition, it can also make the prediction results closer to the user's behavior preferences and conform to the geographical characteristics of the user's location, thereby improving the accuracy of keyword prediction.

[0190] In one embodiment, the prompt template further includes a task output format. Then, the method for generating search guiding suggestion information based on the search expansion words and the guiding suggestions corresponding to the search expansion words in step 1032 can be implemented by at least one of the following methods:

[0191] 301. The server generates and outputs search description information, at least one guiding word, and at least one screening item according to the task output format, based on the search expansion words and the guiding suggestions corresponding to the search expansion words.

[0192] 302. The server generates and outputs search description information and at least one guiding word according to the task output format, based on the search expansion words and the guiding suggestions corresponding to the search expansion words. Then, based on the guiding words, through text matching with the preset product names and / or product categories and / or merchant categories, at least one screening item is obtained.

[0193] Specifically, the server can directly generate search guidance suggestion information according to the task output format in the prompt template and send it to the client, or output the basic content in the search guidance suggestion information, encapsulate it in a preset layout format and then send it to the client, or combine the two information output methods, that is, first output the content generated by the large language model, and then output the typeset search guidance suggestion information to the client according to the preset layout format. This embodiment does not make specific restrictions here.

[0194] In the above embodiment, outputting search guidance suggestion information through the task output format and sending it to the client can improve the speed at which users view the guidance suggestion information; by matching the guiding words with the preset product names and / or product categories and / or merchant categories and generating filtering options, the matching degree between the filtering options and the search keywords can be improved, thereby increasing the relevance between the search results and the search content, and further enhancing the efficiency and accuracy of information search. In this embodiment, the server can first output the search description information, guiding words and other pre-generated content, and then output the filtering options, without having to wait until all the information is generated before sending.

[0195] In one embodiment, the method of generating filtering options from guiding words in step 302 can be implemented by at least one of the following methods:

[0196] 401. The server uses the product names and / or product categories and / or merchant categories that match the guiding words as filtering options.

[0197] 402. The server performs text matching on the product names and / or product categories and / or merchant categories that match the guiding words with the names corresponding to the preset icons, and uses the product names and / or product categories and / or merchant categories, as well as the icons that match the product names and / or product categories and / or merchant categories, as filtering options.

[0198] Specifically, when generating filtering options from guiding words, the server can directly use the product names and / or product categories and / or merchant categories that match the guiding words as filtering options and send them to the client for display; it can also use the product names and / or product categories and / or merchant categories that match the guiding words, as well as the icons corresponding to the product names and / or product categories and / or merchant categories, as filtering options and send them to the client for display.

[0199] For example, Figure 5As shown, after performing text matching on the guiding term "chicken soup" using a preset product name and / or product category name, the screening option "old hen soup" can be obtained. By performing text matching on the name corresponding to the preset icon with the screening option "old hen soup", the icon corresponding to "old hen soup" is not obtained. At this time, "old hen soup" can be directly sent as a screening option to the client for display; conversely, if the icon corresponding to "old hen soup" is obtained, then "old hen soup" and the icon corresponding to "old hen soup" can be jointly sent as a screening option to the client for display.

[0200] In this embodiment, by using the product name and / or product category and / or merchant category that match the guiding term as screening options, it is convenient for users to directly use the screening options as search keywords and thus obtain relatively accurate search results, thereby improving the efficiency and accuracy of information search. Moreover, by using the icons corresponding to the product name and / or product category and / or merchant category as part of the screening options, the richness and aesthetics of the information can be improved. In addition, compared with directly generating the content in the screening options through a generative large model, obtaining the content in the screening options through text matching can make the content in the screening options more matched with the search keywords set in the search engine, and thus it is easier to obtain accurate search results.

[0201] In one embodiment, the prompt templates input into the generative large language model may include a first prompt template and a second prompt template. On this basis, step 103 can be implemented in the following manner: when the number of search results corresponding to the search content is less than the preset quantity threshold, the server generates search guidance suggestion information through the first prompt template; and when the number of search results corresponding to the search content is greater than or equal to the preset quantity threshold, the server generates search guidance suggestion information through the second prompt template.

[0202] In the above embodiments, the content modules included in the first prompt template and the second prompt template may be the same or different. For example, both prompt templates may include at least one of the content modules such as preference dimension, execution steps corresponding to the behavior preference dimension, task objective, execution steps of the task objective, and task output format. At the same time, for the two different search scenarios of zero or few results and broad search terms, the search strategies in the prompt templates may have different focuses, that is, the specific content in the content modules of the two prompt modules may be different. For example, for the search suggestion in the prompt template, when the number of search results corresponding to the search content is less than the preset quantity threshold, the corresponding content in the first prompt template may be "Search suggestion: Provide search ideas around the search keyword"; when the number of search results corresponding to the search content is greater than or equal to the preset quantity threshold, the corresponding content in the second prompt template may be "Search suggestion: Reason for the suggestion, briefly describe the reason in one sentence". In this way, two different prompt templates with different focuses and similar operation methods can be generated.

[0203] In the above embodiments, by providing two different prompt templates when the number of search results corresponding to the search content is greater than or less than a certain quantity threshold, the generated search guidance suggestion information can be more closely combined with the actual search scenario, thereby improving the accuracy and stability of the search guidance suggestion information.

[0204] In one embodiment, the method of displaying the search guidance suggestion information in step 105 can be implemented in the following manner:

[0205] 1051. The client displays at least one of the search description information, guiding words, and filtering options in a structured manner, where the search description information and / or guiding words are displayed in structured text, and the filtering options are displayed in structured controls.

[0206] Specifically, after receiving the search guidance suggestion information sent by the server, the client can display at least one of the search description information, guiding words, and filtering options in a structured manner at different positions on the search result page of the client. Among them, the search description information and / or guiding words can be displayed in structured text, and the filtering options can be displayed in structured controls. In this embodiment, the structured manner refers to the manner of displaying information according to a preset layout manner. The structured display manner is different from the information display manner in the natural state, and can make the displayed information more regular and tidy. And limited by the structured layout manner, the area occupied by each piece of information is limited, so it will not occupy too much space and will not affect other operations such as the user's normal viewing of search results.

[0207] The above embodiments display search guidance suggestion information in a structured manner, which can make the search guidance suggestion information be neatly displayed on the client side, thereby improving the efficiency and comfort of users viewing the search guidance suggestion information, and thus enhancing the search experience. In addition, the search guidance suggestion information displayed on the client side can effectively make up for the problem of low hit rate of existing retrieval results, thereby improving the accuracy of information search.

[0208] In one embodiment, the method of displaying screening options through structured controls in step 1051 can be implemented in the following way: The client displays at least one screening option through sequentially arranged controls, where each control in the sequentially arranged controls is used to display a screening option and is used to respond to a trigger operation to display search results as search results corresponding to the content in the triggered screening option.

[0209] For example, referring to Figure 5 , the client can display multiple screening options through a plurality of sequentially arranged controls, such as "old hen soup", "boiled green vegetables", and "scrambled eggs with tomatoes". Among them, each screening option can respond to a trigger operation to display search results as search results corresponding to the content in the triggered screening option. For example, when the user clicks on the control corresponding to "scrambled eggs with tomatoes", the search result page can display search results corresponding to the keyword "scrambled eggs with tomatoes". It can be understood that the sequential layout referred to in this embodiment can be horizontal arrangement, vertical arrangement, etc. of multiple controls, without specific limitation.

[0210] This embodiment displays multiple screening options through a plurality of sequentially arranged controls, which can quickly clarify the user's search intention and does not require the user to re-enter the search content, thereby effectively improving the user's information search efficiency and improving the user's information search experience.

[0211] In one embodiment, the method of displaying screening options through structured controls in step 1051 can also be implemented in the following way: The client displays the text and / or icon corresponding to the screening option through a control.

[0212] Specifically, referring to Figure 5 , the client can display the text corresponding to the screening option, the icon, or a combination of text and icon through a control. It can be understood that in order to improve the neatness of information display, the information display methods of multiple controls can be the same. For example, multiple controls can only display the text content corresponding to "old hen soup", "boiled green vegetables", and "scrambled eggs with tomatoes" to ensure the neatness of information display.

[0213] The above embodiments display the content corresponding to the screening options in the form of text and / or icons, which can enable users to more easily notice the screening options, thereby facilitating the users to initiate an information search request by triggering the screening options, so as to improve the efficiency of information search.

[0214] In one embodiment, the screening options are obtained by performing text matching on a preset product name and / or product category and / or merchant category based on a guiding word.

[0215] Specifically, the product name and / or product category and / or merchant category can be pre-stored in a screening option vocabulary, or can be stored in their respective vocabularies according to the original storage method. When generating the screening options, by performing text matching one by one between the guiding word and the names in the screening option vocabulary or the existing vocabularies, keywords with a relatively high degree of relevance to the guiding word can be matched as the screening options. In this way, a strong correlation can be established between the guiding word and the screening options, thereby improving the consistency of the search guiding information and the reading experience of users.

[0216] The above embodiments obtain the screening options by performing text matching between the guiding word and a preset product name and / or product category and / or merchant category, which can improve the matching degree between the screening options and the search engine, so as to facilitate users to accurately search for the search results corresponding to the screening options, thereby enhancing the search accuracy and search efficiency of users.

[0217] In one embodiment, the screening options can be a product name and / or product category and / or merchant category that matches the guiding word, and can also be a product name and / or product category and / or merchant category that matches the guiding word, as well as an icon that matches the product name and / or product category and / or merchant category.

[0218] Specifically, after matching the screening options through the product name and / or product category and / or merchant category, the name of the icon in the preset icon library can be matched based on the matched keywords to obtain the icon that matches the screening options, and then the matched product name and / or product category and / or merchant category and the corresponding icon are used as the screening options; in addition, in the case where the icon corresponding to the screening option is not searched / found, the matched product name and / or product category and / or merchant category can also be directly used as the screening options.

[0219] The above embodiments use the product name and / or product category and / or merchant category that matches the guiding word as the screening options, and / or use the product name and / or product category and / or merchant category that matches the guiding word and its icon as the screening options, which can improve the richness of the screening option information, thereby facilitating users to view and trigger the screening options to obtain the search results corresponding to the screening options and improving the search efficiency.

[0220] In one embodiment, the method of displaying search guidance suggestion information in step 105 can also be implemented in the following manner:

[0221] 1052. The client displays search description information and / or guiding words on the first layer of the search result page; and / or, the client displays filtering options and search results on the second layer of the search result page.

[0222] Among them, the first layer and the second layer of the search result page can be two overlapping page layers of the search result page. For example, the first layer is the background layer, and the second layer is the foreground layer that can be slid and displayed on the first layer, or the second layer is the background layer, and the first layer is the foreground layer that can be slid and displayed on the second layer. In addition, the first layer and the second layer can also be two non-overlapping display areas of the search result page, where the placement positions of the first layer and the second layer can be set according to the actual situation. By displaying the search description information and guiding words in one page area, the text content can be relatively concentrated, thereby improving the efficiency of users viewing the search guidance suggestion information; by displaying the filtering options and search results in one page area, it is convenient for users to view the search results corresponding to the information search request, and at the same time, it is convenient for users to trigger the filtering options to refresh the content in the search results, thereby improving the efficiency of users viewing the search results and the accuracy of information search.

[0223] For example, as Figure 5 shown, the client can display search description information such as "Consider using mild foods when having a cold", "Rich in protein and amino acids, improving immunity" and guiding words such as "chicken soup" on the background layer of the search result page, and then display filtering options such as "old hen soup" and search results on the foreground layer of the search result page. By viewing the search description information, users can clarify the next search scope; by viewing the guiding words, users can clarify the next search intention; by clicking on the filtering options, users can directly view the search results corresponding to the filtering options, thereby directly entering the refined search scenario. For example, when the user clicks on the filtering option "scrambled eggs with tomatoes", the client can directly display the search content corresponding to "scrambled eggs with tomatoes" on the foreground layer without having to initiate a new information search request.

[0224] The above embodiment can improve the flexibility of information interaction by displaying various types of information in different areas of the search result page, and can avoid the occlusion of the search guidance suggestion information on the search results, thereby improving the efficiency of users reading the search guidance suggestion information and search results, and enhancing the efficiency of information search and the interactive experience.

[0225] In one embodiment, the search instruction information may include search suggestions and guiding suggestions, where the guiding suggestions are associated with guiding words. On this basis, after step 1052, the information search method may further include at least one of the following implementation manners:

[0226] 1053. In response to an upward sliding operation on the second layer of the search result page, the client hides some of the guiding suggestions and / or some of the guiding words, and when the second layer is pinned to the top for display, hides all of the guiding suggestions and / or all of the guiding words.

[0227] 1054. In response to a downward sliding operation on the second layer of the search result page, the client displays the search instruction information and / or the guiding words, and displays at least one historical search content and the search guiding suggestion information corresponding to the historical search content.

[0228] Specifically, referring to Figure 5 , the client can display search suggestions, guiding words, and the guiding suggestions of the guiding words on the background layer of the search result page, and display filtering options and search results on the foreground layer of the search result page. Through this display method, the user can hide some or all of the guiding words and / or guiding suggestions during the process of swiping up the foreground page, so as to increase the display area of the foreground page and improve the efficiency of the user reading the search results. In this scenario, the search suggestions are not blocked, so the user can still view the search suggestions; during the process of the user swiping down the foreground layer, the user can view the complete search instruction information and guiding words. In addition, the user can also view the historical search content and the search guiding suggestion information corresponding to the historical search content, so as to quickly recall the historical search records and improve the information reading efficiency.

[0229] By controlling the up and down sliding of the second layer of the search result page in the above embodiment, the client can present different display effects, so that the user can flexibly view various information through his own operations, avoiding the search guiding suggestion information occupying too much space, and improving the efficiency and flexibility of the user viewing the search guiding suggestion information and search results.

[0230] In one embodiment, the search instruction information may include search suggestions and guiding suggestions, where the guiding suggestions are associated with guiding words. On this basis, the method for displaying the search guiding suggestion information in step 105 may also be implemented in the following manner: The client receives search suggestions and guiding suggestions in real time and dynamically displays the search suggestions and guiding suggestions; after the search suggestions and guiding suggestions are displayed, each guiding word is used as the beginning of a paragraph, and each guiding word and the guiding suggestion corresponding to the guiding word are displayed in paragraphs.

[0231] Specifically, during the process of the generative large language model outputting content, the server can send the content generated by the large model to the client word by word in real time, and the client can dynamically display the search guidance suggestion information word by word in a typing manner. In this way, the speed of information display can be improved, and the dynamic display method also helps users notice the generated search guidance suggestion information, thereby enhancing the perception of the search guidance suggestion information and further improving the user experience.

[0232] Furthermore, referring to Figure 4 , the client can first display search suggestions and guidance suggestions, such as "When having a cold, you can consider eating mild foods", "Chicken soup is rich in protein and amino acids, which can improve immunity and anti-inflammatory effects; green vegetables are helpful for enhancing immunity", etc., and then use guiding words such as "chicken soup", "green vegetables", and "tomato" as the beginning of paragraphs to display each guiding word and the guiding suggestions corresponding to the guiding words in paragraphs. In this way, on the one hand, the effect of dynamic display can be enhanced, and the user's perception can be improved; on the other hand, the comfort of users viewing the text can be improved. Also, limited by the text length after segmentation, the area where the displayed information is located is limited and will not affect other operations of users viewing search results, etc. In addition, the search suggestions, guiding words, and guiding suggestions displayed by the client can effectively make up for the problem of low hit rate of existing retrieval results, thereby improving the accuracy of information search.

[0233] In one embodiment, the information search method may further include the following steps:

[0234] 106. After the search guidance suggestion information is displayed on the client, control the device where the client is located to vibrate and prompt.

[0235] Specifically, after the client completely displays the search guidance suggestion information, it can control the device where the client is located to vibrate and prompt to prompt the user to view the relevant content of artificial intelligence search. In this way, the user's perception and participation in the search guidance suggestion information can be improved. And in this embodiment, by displaying the search guidance suggestion information, the search accuracy can be improved, and the problem of low hit rate of traditional search results can be improved, thereby improving the user's search experience.

[0236] In one embodiment, the above information search method may further include at least one of the following methods:

[0237] 501. Before displaying the search guidance suggestion information on the client, display a prompt message indicating that the search guidance suggestion information is being generated, where the prompt message is generated according to a preset prompt message and / or keywords in the search content.

[0238] Specifically, referring to Figure 2 and Figure 3, before the client displays the search guidance suggestion information, it can display a prompt message indicating that the search guidance suggestion information is being generated, such as "Detecting fewer results, casting a spell...", "Dishes suitable for colds", and so on. Since it takes a certain amount of time for the generative large language model to generate the search guidance suggestion information and send it to the client, displaying the prompt message allows users to understand the relevant functions of artificial intelligence search in advance, thereby enhancing the perception of the search guidance suggestion information.

[0239] 502. Before the client displays the search guidance suggestion information, it dynamically displays an information loading icon.

[0240] Specifically, referring to Figure 2 and Figure 3 , before the client displays the search guidance suggestion information, it can dynamically display an information loading icon. The information loading icon can be displayed together with the prompt message or separately. By dynamically displaying the information loading icon, the perception of the artificial intelligence search function can be enhanced.

[0241] 503. The client displays the search content and an artificial intelligence icon in the information search box.

[0242] Specifically, referring to Figures 2 to 5 , the client can display the search content and an artificial intelligence icon in the information search box, where the artificial intelligence icon is used to identify that the information search request corresponding to the search content is a preset artificial intelligence search request. By displaying the artificial intelligence icon, it can prompt the user that the initiated information search request is a preset artificial intelligence search request, thereby enhancing the user's perception of the artificial intelligence search function.

[0243] 504. In response to detecting a network interruption, the client displays an information loading control, where the information loading control is used to, in response to a trigger operation, re-receive and display the search guidance suggestion information.

[0244] Specifically, after the information search request is sent, the client may interrupt the process of receiving the search guidance suggestion information due to network problems during the process of receiving the search guidance suggestion information. The client can detect the information reception status or network status in real time. If a network interruption or information reception interruption is detected, it can display the information loading control. After the user clicks the information loading control, the client can re-receive the search guidance suggestion information sent by the server, where the reception method can be receiving from the beginning or receiving intermittently, and this embodiment does not make specific restrictions.

[0245] Through the above-mentioned embodiments, through various information display methods, the user's perception and participation in the artificial intelligence search function can be enhanced, thereby enhancing the user's search experience.

[0246] In one embodiment, as a refinement and extension of the specific implementation manner of the above embodiment, to fully illustrate the implementation process of this embodiment, an information search method is provided. As shown in Figure 6 and Figure 7 , this method can be implemented in the following way: When the user enters search content (at this time, it is the pre-search content) in the information search box of the client, the client can respond to the information input operation, display the search association page, and send the pre-search content to the server. The server can identify the keywords in the pre-search content and determine whether the keywords in the pre-search content match the keywords in the artificial intelligence offline word list. If the keywords in the word list are hit, the artificial intelligence entry can be encapsulated through the artificial intelligence entry and the artificial intelligence mode marker on the server side, and the artificial intelligence entry can be sent to the client. The client can display the artificial intelligence entry in the search association page. At the same time, other relevant entries related to the keywords in the search content can also be displayed. At this time, the client can execute different search methods according to different operations performed by the user.

[0247] Furthermore, if the user directly sends the search content and the search content does not meet the preset conditions, or selects a common entry for sending, the client can send the information search request to the search engine of the server to obtain the search results corresponding to the information search request, and display the search results on the search result page; if the user sends the search content and the search content meets the preset conditions, or clicks on the artificial intelligence entry, on the one hand, the client can send the information search information search request to the search engine of the server to obtain the search results corresponding to the information search request, and then display the search results on the search result page; on the other hand, the artificial intelligence entry and the information search request can be sent to the generative large language model of the server. Then, the server can start the generative large language model. The large model can select the corresponding prompt template according to the number of search results corresponding to the search content and perform information prediction such as expanding keywords, so as to generate search guidance suggestion information. Then, after encapsulating the generated search guidance suggestion information, it is sent to the client for rendering and display. At this time, the user can view the search guidance suggestion information corresponding to the artificial intelligence entry through the client, and at the same time, can also view the search results corresponding to the search content.

[0248] The above method can automatically trigger the preset artificial intelligence search request by responding to the user's search operation, which can effectively improve the user's search efficiency and the accuracy of the search results. At the same time, the above method will not affect the display of normal search results, will not cause too much interference to the user, and can make the search results closer to the user's actual search needs, thereby effectively improving the user's search experience.

[0249] 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 for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. In addition, the labels corresponding to the steps in the above embodiments only serve as identification and do not limit the execution order of the steps. The execution order of the steps in each embodiment can be set according to the actual situation.

[0250] Furthermore, as Figures 1 to 7 a specific implementation of the method shown, this embodiment provides an information search device, as Figure 8 shown. The device includes: a search request response module 61 and a guidance suggestion display module 62, where:

[0251] The search request response module 61 is configured to, in response to receiving an information search request, send the information search request;

[0252] The guidance suggestion display module 62 is configured to receive and display search guidance suggestion information and search results. Among them, the search guidance suggestion information is generated by a generative large language model based on the search content, and the search results are obtained through the information search request; the search guidance suggestion information includes at least one of search description information, guiding words, and filtering options; the guiding words and the filtering options are mutually related, and the filtering options are used to display the search results as search results corresponding to the content in the filtering options.

[0253] In a specific application scenario, the guidance suggestion display module 62 is specifically configured to: display at least one of the search description information, the guiding words, and the filtering options in a structured manner, where the search description information and / or the guiding words are displayed in structured text, and the filtering options are displayed in structured controls.

[0254] In a specific application scenario, the guidance suggestion display module 62 is specifically configured to: display at least one of the filtering options through sequentially arranged controls, where each control in the sequentially arranged controls is used to display one of the filtering options and is configured to, in response to a trigger operation, display the search results as search results corresponding to the content in the triggered filtering option.

[0255] In a specific application scenario, the guidance suggestion display module 62 is specifically configured to: display the text and / or icon corresponding to the filtering option through the control.

[0256] In a specific application scenario, the guidance and suggestion display module 62 is specifically configured to: display the search description information and / or the guiding words on the first layer of the search result page; and / or display the filtering options and the search results on the second layer of the search result page.

[0257] In a specific application scenario, the search description information includes search suggestions and guiding suggestions, and the guiding suggestions are associated with the guiding words; the guiding suggestion display module 62 is further configured to: in response to an upward sliding operation on the second layer of the search result page, hide some of the guiding suggestions and / or some of the guiding words, and when the second layer is pinned to the top for display, hide all of the guiding suggestions and / or all of the guiding words; and / or in response to a downward sliding operation on the second layer of the search result page, display the search description information and / or the guiding words, and display at least one historical search content and the search guiding suggestion information corresponding to the historical search content.

[0258] In a specific application scenario, the filtering options are obtained by performing text matching on a preset product name and / or product category and / or merchant category based on the guiding words.

[0259] In a specific application scenario, the filtering options are product names and / or product categories and / or merchant categories that match the guiding words; and / or the filtering options are product names and / or product categories and / or merchant categories that match the guiding words, and icons that match the product names and / or product categories and / or merchant categories.

[0260] In a specific application scenario, the search request response module 61 is specifically configured to: in response to detecting that an information search request initiated by a user is a preset artificial intelligence search request, send the information search request.

[0261] In a specific application scenario, the search request response module 61 is specifically configured to: in response to a triggering operation on an artificial intelligence search entry and / or an artificial intelligence entry, determine that an information search request initiated by a user is a preset artificial intelligence search request; and / or in the artificial intelligence mode, in response to a sending operation of the search content, determine that an information search request initiated by a user is a preset artificial intelligence search request; and / or in the case where the search content input by the user meets a preset condition, in response to a sending operation of the search content, determine that an information search request initiated by a user is a preset artificial intelligence search request, where the preset condition includes that a keyword in the search content matches a preset keyword, and / or the number of search results corresponding to the search content is less than a preset quantity threshold.

[0262] In a specific application scenario, the search request response module 61 is further configured to: in response to receiving the pre-search content input by the user, send the pre-search content; when the keyword in the pre-search content matches the preset keyword, and / or the number of search results corresponding to the pre-search content is less than the preset quantity threshold, display the artificial intelligence search entry and / or the artificial intelligence entry, where the artificial intelligence entry is determined according to the keyword in the pre-search content.

[0263] In a specific application scenario, the search request response module 61 is further configured to: in response to the trigger operation on the artificial intelligence search entry and / or the artificial intelligence entry, enter the artificial intelligence mode; and / or when the search content input by the user meets the preset conditions, in response to at least one of the send operation, modification operation, and replacement operation of the search content, enter the artificial intelligence mode.

[0264] In a specific application scenario, the search instruction information includes search suggestions and guiding suggestions, and the guiding suggestions are associated with the guiding words; the guiding suggestion display module 62 is further specifically configured to: receive the search suggestions and the guiding suggestions in real time and dynamically display the search suggestions and the guiding suggestions; after the search suggestions and the guiding suggestions are displayed, use each of the guiding words as the beginning of a paragraph and display each guiding word and the guiding suggestion corresponding to the guiding word in a segmented manner.

[0265] In a specific application scenario, the guiding suggestion display module 62 is further configured to: after the search guiding suggestion information is displayed, control the device where the client is located to perform a vibration prompt.

[0266] In a specific application scenario, the guiding suggestion display module 62 is further configured to: before displaying the search guiding suggestion information, display a prompt information indicating that the search guiding suggestion information is being generated, where the prompt information is generated according to the preset prompt information and / or the keyword in the search content; and / or before displaying the search guiding suggestion information, dynamically display an information loading icon; and / or display the search content and an artificial intelligence icon in the information search box, where the artificial intelligence icon is used to identify that the information search request corresponding to the search content is a preset artificial intelligence search request; and / or in response to detecting a network interruption, display an information loading control, where the information loading control is used to, in response to a trigger operation, re-receive and display the search guiding suggestion information.

[0267] It should be noted that for other corresponding descriptions of each functional unit involved in the information search device provided in this embodiment, reference can be made to Figures 1 to 7The corresponding description in [reference] will not be elaborated here.

[0268] Further, as Figures 1 to 7 a specific implementation of the method shown, this embodiment provides an information search device, as Figure 9 shown, the device includes: a search request response module 71, a guidance suggestion generation module 72, and a guidance suggestion sending module 73.

[0269] The search request response module 71 is configured to, in response to receiving an information search request, obtain the search content corresponding to the information request;

[0270] The guidance suggestion generation module 72 is configured to, based on the search content, generate search guidance suggestion information through a generative large language model, and obtain a search result corresponding to the information search request, where the search guidance suggestion information includes at least one of search description information, guiding words, and filtering options, and the guiding words and the filtering options are correlated with each other;

[0271] The guidance suggestion sending module 73 is configured to send the search guidance suggestion information and the search result to the client, and the filtering option is configured to display the search result as a search result corresponding to the content in the filtering option.

[0272] In a specific application scenario, the filtering option is obtained by performing text matching on a preset product name and / or product category and / or merchant category based on the guiding words.

[0273] In a specific application scenario, the filtering option is a product name and / or product category and / or merchant category that matches the guiding words; and / or the filtering option is a product name and / or product category and / or merchant category that matches the guiding words, and an icon that matches the product name and / or product category and / or merchant category.

[0274] In a specific application scenario, the search request response module 71 is further configured to: in response to receiving pre-search content input by a user, determine whether a keyword in the pre-search content matches a preset keyword, and / or determine whether the number of search results corresponding to the pre-search content is less than a preset quantity threshold; when the keyword in the pre-search content matches the preset keyword, and / or the number of search results corresponding to the pre-search content is less than the preset quantity threshold, send an artificial intelligence search entry and / or an artificial intelligence entry to the client, where the artificial intelligence entry is determined according to the keyword in the pre-search content.

[0275] In a specific application scenario, the guidance and suggestion generation module 72 is specifically configured to: input the search content, the user information initiating the information search request, and a preset prompt template into the generative large language model, so that the generative large language model generates user behavior preference information based on the prompt template and the user information, and generates the search guidance and suggestion information based on the prompt template, the user behavior preference information, and the search content.

[0276] In a specific application scenario, the prompt template includes a first prompt template and a second prompt template; the guidance and suggestion generation module 72 is further specifically configured to: when the number of search results corresponding to the search content is less than a preset number threshold, generate the search guidance and suggestion information through the first prompt template; when the number of search results corresponding to the search content is greater than or equal to the preset number threshold, generate the search guidance and suggestion information through the second prompt template.

[0277] In a specific application scenario, the prompt template includes at least one behavior preference dimension and the execution steps corresponding to the behavior preference dimension; the guidance and suggestion generation module 72 is specifically configured to: through the execution steps corresponding to the behavior preference dimension, generate user behavior preference information of at least one of the behavior preference dimensions based on the user information, where the behavior preference dimensions of the user behavior preference information include at least one of keyword preference, product category preference, merchant brand preference, delivery speed preference, order placement time preference, specific product preference, and product attribute preference.

[0278] In a specific application scenario, the prompt template further includes a task objective and the execution steps of the task objective; the guidance and suggestion generation module 72 is specifically configured to: through the task objective and the execution steps of the task objective, construct a semantic expansion network based on the keywords in the search content and the related words corresponding to the keywords, and predict at least one search expansion word and the guidance suggestions corresponding to the search expansion word based on the user behavior preference information and / or the user location information through the semantic expansion network; generate the search guidance and suggestion information based on the search expansion word and the guidance suggestions corresponding to the search expansion word.

[0279] In a specific application scenario, the guidance suggestion generation module 72 is specifically configured to: generate and output the search description information, at least one of the guidance words, and at least one of the screening options according to the task output format, based on the search expansion words and the guidance suggestions corresponding to the search expansion words; and / or generate and output the search description information and at least one of the guidance words according to the task output format, based on the search expansion words and the guidance suggestions corresponding to the search expansion words; and obtain at least one screening option by performing text matching on the guidance words through a preset product name and / or product category and / or merchant category.

[0280] It should be noted that for other corresponding descriptions of each functional unit involved in the information search device provided in this embodiment, reference can be made to Figures 1 to 7 the corresponding description in, which will not be elaborated here.

[0281] Based on the above as Figures 1 to 7 shown in the method, correspondingly, this embodiment also provides a storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the information search method as Figures 1 to 7 shown above.

[0282] Based on such an understanding, the technical solution of this application can be embodied in the form of a software product. The software product to be recognized can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation scenarios of this application.

[0283] Based on the above as Figures 1 to 7 shown in the method, and Figure 8 and Figure 9 shown in the information search device embodiment, in order to achieve the above purpose, as Figure 10 shown, this embodiment also provides a computer device for information search, which can specifically be a personal computer, a server, a smart phone, a tablet computer, a smart watch, or other network devices, etc. The computer device includes a storage medium and a processor; the storage medium is used to store a computer program and an operating system; the processor is used to execute the computer program to implement the method as Figures 1 to 7 shown above.

[0284] Optionally, the computer device may further include an internal memory, a communication interface, a network interface, a camera, a Radio Frequency (RF) circuit, sensors, an audio circuit, a WI-FI module, a display screen (Display), an input device such as a keyboard (Keyboard), etc. Optionally, the communication interface may further include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface), etc.

[0285] Those skilled in the art can understand that the structure of the computer device for identifying an operation action provided in this embodiment does not limit the computer device, and it may include more or fewer components, or combine certain components, or have different component arrangements.

[0286] The storage medium may further include an operating system and a network communication module. The operating system is a program for managing the hardware and software resources to be identified of the above computer device, and supports the operation of the information processing program and other software and / or programs to be identified. The network communication module is used to implement communication between various components inside the storage medium, as well as communication between other hardware and software in the information processing computer device.

[0287] Through the description of the above embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus a necessary general hardware platform, or by hardware. By applying the technical solution of this application, first, in response to an information search request initiated by a user, the information search request is sent to the server so that the server obtains the search content corresponding to the information search request. Then, according to the search content, through a generative large language model, search guidance suggestion information covering various information such as search description information, guiding words, and screening options is generated, and the search result corresponding to the information search request is obtained. Finally, the generated search guidance suggestion information and the obtained search result are sent to the client for display. Compared with the prior art, the above method can provide more accurate and personalized search results for users, and present the search results to users in a more interpretable manner, thereby improving the accuracy of search results and enhancing the user's information search experience.

[0288] Those skilled in the art can understand that the drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the drawings are not necessarily essential for implementing this application. Those skilled in the art can understand that the modules in the device in the implementation scenario can be distributed in the device in the implementation scenario according to the description of the implementation scenario, or can be correspondingly changed to be located in one or more devices different from this implementation scenario. The modules in the above implementation scenario can be combined into one module, or further split into multiple sub-modules.

[0289] The above serial numbers of the present application are only for description and do not represent the advantages or disadvantages of the implementation scenarios. The above disclosure only shows several specific implementation scenarios of the present application. However, the present application is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the protection scope of the present application.

Claims

1. An information search method, characterized in that, The method includes: Upon receiving an information search request, the client sends the information search request to the server; Upon receiving the information search request, the server obtains the search content corresponding to the information search request; Based on the search content, the server generates search guidance suggestion information through a generative large language model and obtains search results corresponding to the information search request, where the search guidance suggestion information includes at least one of search description information, guiding words, and filtering options, and the guiding words and the filtering options are interrelated; The server sends the search guidance suggestion information and the search results to the client; The client receives and displays the search guidance suggestion information and the search results, and the filtering options are used to display the search results as search results corresponding to the content in the filtering options.

2. The method according to claim 1, wherein The client receiving and displaying the search guidance suggestion information includes: The client displays at least one of the search description information, the guiding words, and the filtering options in a structured manner, where the search description information and / or the guiding words are displayed in structured text, and the filtering options are displayed in structured controls.

3. The method according to claim 2, wherein The client displaying the filtering options through the structured controls includes: The client displays at least one of the filtering options through sequentially arranged controls, where each control in the sequentially arranged controls is used to display one of the filtering options and is used to, in response to a trigger operation, display the search results as search results corresponding to the content in the triggered filtering option.

4. The method according to claim 2 or 3, characterized in that The client displaying the filtering options through the structured controls further includes: The client displays the text and / or icon corresponding to the filtering options through the controls.

5. An information search method, characterized in that, The method includes: Upon receiving an information search request, sending the information search request; Receiving and displaying search guidance suggestion information and search results, where the search guidance suggestion information is generated through a generative large language model based on the search content, and the search results are obtained through the information search request; the search guidance suggestion information includes at least one of search description information, guiding words, and filtering options; the guiding words and the filtering options are interrelated, and the filtering options are used to display the search results as search results corresponding to the content in the filtering options.

6. An information search method, characterized in that, The method includes: Upon receiving an information search request, obtaining the search content corresponding to the information request; Based on the search content, generating search guidance suggestion information through a generative large language model and obtaining search results corresponding to the information search request, where the search guidance suggestion information includes at least one of search description information, guiding words, and filtering options, and the guiding words and the filtering options are interrelated; Sending the search guidance suggestion information and the search results to the client, and the filtering options are used to display the search results as search results corresponding to the content in the filtering options.

7. An information search device, characterized in that, The device includes: A search request response module, configured to send the information search request in response to receiving an information search request; A guidance suggestion display module, configured to receive and display search guidance suggestion information and search results, wherein the search guidance suggestion information is generated by a generative large language model based on the search content, and the search results are obtained through the information search request; the search guidance suggestion information includes at least one of search description information, guiding words, and filtering options; the guiding words and the filtering options are correlated with each other, and the filtering options are used to display the search results as search results corresponding to the content in the filtering options.

8. An information search device, characterized in that, The device includes: A search request response module, configured to obtain the search content corresponding to the information request in response to receiving an information search request; A guidance suggestion generation module, configured to generate search guidance suggestion information through a generative large language model based on the search content, and obtain search results corresponding to the information search request, wherein the search guidance suggestion information includes at least one of search description information, guiding words, and filtering options, and the guiding words and the filtering options are correlated with each other; A guidance suggestion sending module, configured to send the search guidance suggestion information and the search results to the client, and the filtering options are used to display the search results as search results corresponding to the content in the filtering options.

9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images