Object recommendation method and device

Abstract

The embodiment of the invention provides an object recommendation method and device.The object recommendation method is applied to an intelligent processing unit and comprises the steps that object query information sent by a target account is obtained, and multiple to-be-recommended objects associated with the object query information are determined in an object library; sending an information consultation request to the object providing ends corresponding to the plurality of to-be-recommended objects, and screening out at least one recommended object from the plurality of to-be-recommended objects based on a consultation result; and generating an object recommendation report based on the object query information, the consultation result and the object detail information of the at least one recommendation object.

Description

Technical Field

[0001] This specification relates to the field of computer technology, and in particular to object recommendation methods and apparatus. Background Technology

[0002] With the rapid development of internet technology and the widespread adoption of e-commerce platforms, online shopping has gradually become a mainstream consumption method for consumers. However, faced with tens of thousands of products on e-commerce platforms, users need to browse each product individually to select those that meet their needs. This consumes a significant amount of time and energy, greatly impacting the user's shopping experience. Summary of the Invention

[0003] In view of this, embodiments of this specification provide an object recommendation method. One or more embodiments of this specification also relate to an object recommendation apparatus, a computing device, a computer-readable storage medium, and a computer program product, to address the technical deficiencies existing in the prior art.

[0004] According to a first aspect of the embodiments of this specification, an object recommendation method is provided, applied to an intelligent processing unit, comprising: Obtain object query information sent by the target account, and identify multiple objects to be recommended in the object database that are associated with the object query information; Send information consultation requests to the object providers corresponding to the plurality of objects to be recommended, and select at least one recommended object from the plurality of objects to be recommended based on the consultation results; Based on the object query information, the consultation results, and the object details information of the at least one recommended object, an object recommendation report is generated.

[0005] According to a second aspect of the embodiments of this specification, another object recommendation method is provided, applied to cloud-side devices, including: Obtain object query information sent by the terminal device, and determine multiple objects to be recommended in the object database that are associated with the object query information; Send information consultation requests to the object providers corresponding to the plurality of objects to be recommended, and select at least one recommended object from the plurality of objects to be recommended based on the consultation results; Based on the object query information, the consultation results, and the object details information of the at least one recommended object, an object recommendation report is generated and sent to the end device.

[0006] According to a third aspect of the embodiments of this specification, an object recommendation apparatus is provided, applied to an intelligent processing unit, comprising: The first acquisition module is configured to acquire object query information sent by the target account and determine multiple objects to be recommended in the object library that are associated with the object query information; The first sending module is configured to send information consultation requests to the object providers corresponding to the plurality of objects to be recommended, and to select at least one recommended object from the plurality of objects to be recommended based on the consultation results. The first generation module is configured to generate an object recommendation report based on the object query information, the consultation results, and the object details information of the at least one recommended object.

[0007] According to a fourth aspect of the embodiments of this specification, another object recommendation apparatus is provided, applied to cloud-side devices, comprising: The second acquisition module is configured to acquire object query information sent by the end device and determine multiple objects to be recommended in the object library that are associated with the object query information; The second sending module is configured to send information consultation requests to the object providers corresponding to the plurality of objects to be recommended, and to select at least one recommended object from the plurality of objects to be recommended based on the consultation results. The second generation module is configured to generate an object recommendation report and send it to the end device based on the object query information, the consultation results, and the object details information of the at least one recommended object.

[0008] According to a fifth aspect of the embodiments of this specification, a computing device is provided, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of any of the above-described object recommendation methods.

[0009] According to a sixth aspect of the embodiments of this specification, a computer-readable storage medium is provided that stores computer-executable instructions, which, when executed by a processor, implement the steps of any of the above-described object recommendation methods.

[0010] According to a seventh aspect of the embodiments of this specification, a computer program product is provided, including a computer program or instructions that, when executed by a processor, implement the steps of any of the above-described object recommendation methods.

[0011] The object recommendation method provided in this embodiment obtains object query information sent by a target account, identifies multiple objects to be recommended in the object database associated with the object query information, and sends information consultation requests to the object providers corresponding to each of the multiple objects to be recommended, greatly enriching the information dimensions of the objects. Then, based on the consultation results, at least one recommended object is selected from the multiple objects to be recommended, making the recommended objects more accurately meet the user's needs. Finally, based on the object query information, consultation results, and detailed information of the object corresponding to at least one recommendation, an object recommendation report is generated, making the recommendation results transparent and well-founded. Using the object recommendation method provided in this embodiment, users only need to provide their needs, without having to search through the object database one by one to obtain an object recommendation report, and then make further decisions based on the object recommendation report, greatly improving the efficiency of users selecting objects in the object database, thereby enhancing the user experience. Attached Figure Description

[0012] Figure 1 This is a schematic diagram of an object recommendation method provided in one embodiment of this specification; Figure 2 This is a flowchart illustrating an object recommendation method provided in one embodiment of this specification; Figure 3 This is a flowchart of another object recommendation method provided in one embodiment of this specification; Figure 4 This is a flowchart illustrating the processing procedure of an object recommendation method provided in one embodiment of this specification; Figure 5 This is a schematic diagram of a product recommendation report provided in one embodiment of this specification; Figure 6 This is a schematic diagram of the structure of an object recommendation device provided in one embodiment of this specification; Figure 7 This is a schematic diagram of another object recommendation device provided in one embodiment of this specification; Figure 8 This is a structural block diagram of a computing device provided in one embodiment of this specification. Detailed Implementation

[0013] Many specific details are set forth in the following description to provide a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar extensions without departing from the spirit of this specification. Therefore, this specification is not limited to the specific implementations disclosed below.

[0014] The terminology used in one or more embodiments of this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of this specification. The singular forms “a,” “described,” and “the” as used in one or more embodiments of this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0015] It should be understood that although the terms first, second, etc., may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first may also be referred to as second without departing from the scope of one or more embodiments of this specification, and similarly, second may also be referred to as first. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0016] Furthermore, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties. Moreover, the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0017] The technical solutions provided in this application can employ deep learning models with relatively large parameter scales. However, this large model is merely an example; this application does not limit the number of model parameters supported by the deep learning model used, aiming to meet actual needs. The deep learning models involved in this application can be artificial intelligence-based language models (LM) or multimodal models (MM).

[0018] This specification provides an object recommendation method, and one or more embodiments of this specification also relate to an object recommendation apparatus, a computing device, a computer-readable storage medium, and a computer program product, which will be described in detail in the following embodiments.

[0019] See Figure 1The schematic diagram illustrates that the object recommendation method provided in this embodiment obtains object query information sent by the target account, identifies multiple objects to be recommended in the object library associated with the object query information, and sends information consultation requests to the object providers corresponding to each of the multiple objects to be recommended, greatly enriching the information dimensions of the objects. Then, based on the consultation results, at least one recommended object is selected from the multiple objects to be recommended, making the recommended objects more accurately meet the user's needs. Finally, based on the object query information, consultation results, and detailed information of the object corresponding to at least one recommendation, an object recommendation report is generated, making the recommendation results transparent and well-founded. Using the object recommendation method provided in this embodiment, users only need to provide their needs, without having to search through the object library one by one to obtain an object recommendation report, and then make further decisions based on the object recommendation report, greatly improving the efficiency of users selecting objects from the object library, thereby enhancing the user experience.

[0020] See Figure 2 , Figure 2 A flowchart of an object recommendation method according to an embodiment of this specification is shown, which specifically includes the following steps.

[0021] Step S202: Obtain object query information sent by the target account, and determine multiple objects to be recommended in the object database that are associated with the object query information.

[0022] The method provided in this embodiment can be applied to an intelligent processing unit, which can be an intelligent agent. An intelligent agent is an intelligent program built based on a model. The intelligent agent can perceive information, analyze the information, and finally make a decision based on the analysis results.

[0023] The object recommendation method provided in this embodiment can be applied to any scenario where objects need to be selected from an object library. These objects can be products, services, etc. For example, to obtain a purchase requirement for a certain product, one can consult with multiple suppliers and generate a product recommendation report based on their responses. As another example, if the object is a customized service (such as travel planning, home design, etc.), after obtaining personalized requirements, one can consult with designers or planners to obtain a service recommendation report.

[0024] This embodiment uses the object recommendation method in the product recommendation scenario as an example for illustration. The descriptions of other scenarios can be found in the same or corresponding descriptions in this embodiment, and will not be elaborated on in detail here.

[0025] Specifically, the target account refers to the account of the user who initiated the object recommendation request. Object query information refers to information provided by the target account describing object needs and preferences. The object repository refers to a database storing a large number of objects. The objects to be recommended are those selected from the object repository that match the object query information through matching.

[0026] Based on this, the intelligent processing unit can establish a connection with the object database to query and retrieve relevant information about each object in the database. The intelligent processing unit can obtain object query information sent by the target account. This object query information can describe the user's needs for the object, and may also include information about the target account's preferences for the object obtained from their behavior. After determining the object query information, the unit can identify multiple objects in the object database that match the query information for recommendation, based on the matching degree between the query information and the objects in the database.

[0027] Furthermore, object query information can be obtained based on object requirement information. In this embodiment, the specific implementation is as follows: Receive object requirement information sent by the target account; parse the object requirement information to obtain the object query information corresponding to the object requirement information.

[0028] Specifically, object requirement information refers to the initial requirement description of the object sent by the target account. Object requirement information can include various formats, such as natural language, tables, documents, etc. The specific format can be set by relevant technical personnel, and this embodiment does not impose any limitations.

[0029] Based on this, the intelligent processing unit receives object requirement information sent by the target account. This object requirement information can be in the form of a table or text. It then parses the object requirement information to obtain the corresponding object query information. For example, when the object requirement information is text, a large language model can be used to parse the text, thereby obtaining structured requirement information.

[0030] For example, the customer's needs are to purchase ballpoint pens, requiring black ink, a unit price budget of less than 5 yuan, smooth writing, quick-drying and waterproof, suitable for daily office writing, and an invoice is required.

[0031] In summary, the system receives object request information from the target account, parses it to obtain object query information, and structures the object query information, thereby improving the efficiency of filtering objects to be recommended in the object library.

[0032] Furthermore, if the object requirement information is incomplete, it can be supplemented through prompts. In this embodiment, the specific implementation method is as follows: Receive initial requirement information sent by the target account, compare the initial requirement information with the query template; if the comparison result shows that there is missing requirement information, generate a prompt message based on the missing requirement information and send it to the target account to obtain supplementary requirement information; merge the initial requirement information and the supplementary requirement information to obtain the object requirement information.

[0033] Specifically, initial requirement information refers to the potentially incomplete requirement description information submitted by the target account for the first time. A query template is a predefined set of fields for querying a specific type of object. For example, when the object is a product, the query template could include product category, purchase quantity, budget range, and purchase requirements (logistics, delivery time, service warranty, supplier requirements), etc. Missing requirement information refers to information not included in the initial requirement information but corresponding to fields in the template, determined by comparing the initial requirement information with the query template. Prompt information refers to guiding information sent to the target account to obtain missing requirement information. The format of prompt information can be a question-and-answer format, or an option or fill-in-the-blank format, etc. The specific format can be determined by relevant technical personnel according to the actual situation; this embodiment does not impose any limitations. Supplementary requirement information refers to the information returned by the target account after receiving the prompt information.

[0034] Based on this, the intelligent processing unit receives the initial demand information sent by the target account, and then parses the initial demand information to obtain various attribute information about the product. A large language model can be used to parse the initial demand information. Then, the attribute information is compared with various fields included in the query template to check for missing demand information. If missing demand information exists, guiding prompts are generated based on the missing demand information. For example, a natural language question can be generated using the large language model to address the missing demand information, or the user can be guided to send supplementary information in a multiple-choice or fill-in-the-blank format. The missing supplementary demand information is then obtained based on the supplementary information sent by the user. Finally, the initial demand information and the supplementary demand information are merged to obtain the target demand information.

[0035] For example, suppose the initial requirement is to purchase ballpoint pens with black refills. The prompt message is "What is your budget for a single ballpoint pen? Do you need an invoice?", and the supplementary requirement is a unit price budget of 5 yuan or less, and an invoice is required. After merging, the resulting requirement is the same: to purchase ballpoint pens with black refills, a unit price budget of 5 yuan or less, and an invoice is required.

[0036] In summary, preset query templates can be used to prompt users to supplement missing information, thereby making the obtained object requirement information more comprehensive. This leads to more comprehensive object query information based on the object requirement information, which in turn improves the efficiency of subsequent object filtering in the object library and makes the filtered objects more in line with the user's needs.

[0037] Furthermore, behavioral information of the target account can be obtained, and the target query information can be determined based on the behavioral information and the target's needs. In this embodiment, the specific implementation is as follows: Obtain the behavioral information of the target account; parse the object requirement information and the behavioral information to obtain the object query information corresponding to the object requirement information.

[0038] Specifically, behavioral information refers to the target account's historical activity information. This information can include the target account's browsing history, search history, and interaction history (such as likes, favorites, contacts, and purchases). When the object is a product, the target account can be an account on an e-commerce platform, and the target account's behavioral information can include order information, product and merchant favorites information, product browsing information, product search information, and so on.

[0039] Based on this, the intelligent processing unit can acquire the target account's behavioral information, then parse the object requirement information and behavioral information to obtain object query information. It can identify and extract object-related attribute information from the object requirement information to obtain a set of object attribute information; specifically, a large language model can be used to identify and extract object requirement information. Regarding the target account's behavioral information, user preferences can be obtained based on the behavioral information, such as the target account's historical browsing history, historical search history, and historical interaction history (e.g., likes, favorites, contacts, purchases, etc.). Then, the behavioral information is parsed. Specifically, this involves identifying multiple objects corresponding to the behavioral information, such as browsing objects, purchase objects, favorite objects, etc., and then identifying relevant objects from these multiple objects. These relevant objects are those related to the objects in the object requirement information. Finally, information about these relevant objects in the behavioral information is extracted and summarized. Finally, the parsed object requirement information and behavioral information are combined to obtain object query information.

[0040] Continuing the previous example, retrieve the product order and identify the purchase record for ballpoint pens within that order. The object query information is: based on the purchase record for ballpoint pens and the object's requirement information, "Need to purchase ballpoint pens, the ink color must be black, the unit price budget is within 5 yuan, and an invoice is required."

[0041] In summary, behavioral information can be further obtained, and then object query information can be generated based on behavioral information and object demand information. This allows the products filtered based on object query information to better meet the user's personalized needs.

[0042] Furthermore, multiple objects to be recommended can be identified from the object library based on object demand information and behavioral information. In this embodiment, the specific implementation is as follows: Based on the target account's object requirement information, multiple candidate objects matching the object requirement information are filtered out from the object library; the matching degree scores of the multiple candidate objects with the target account's behavioral information are calculated, and the multiple candidate objects are sorted according to the matching degree scores; a set number of candidate objects are selected from the sorted multiple candidate objects as multiple objects to be recommended.

[0043] Specifically, a candidate object refers to an object in the object database that meets the conditions in the object requirement information. The matching score is a quantitative score used to measure the degree of fit between the behavioral information of the "candidate object" and the "target account".

[0044] Based on this, the intelligent processing unit filters candidate objects that match the object requirements from the object database according to the various requirements for the object in the object requirement information. Then, it further filters the candidate objects based on behavioral information to obtain the objects to be recommended. Specifically, after obtaining the candidate objects, the unit can match the candidate objects with the behavioral information to obtain a matching score. It can compare the various attributes of the candidate objects, score each attribute based on the comparison results, and finally calculate the matching score between the candidate objects and the behavioral information by weighting the scores of each attribute. Finally, the candidate objects are sorted based on the matching score, and a set number of candidate objects with higher matching scores are selected as the objects to be recommended. The specific number selected can be determined by relevant technical personnel according to actual needs, and this embodiment does not impose any limitations.

[0045] For example, based on the object requirement information, candidate objects are filtered in the object library, and the matching score is calculated. The matching scores are arranged in descending order as: object 1, object 2, object 3, object 4... object 100. Suppose 10 candidate objects are selected as the objects to be recommended, and the objects to be recommended are: object 1, object 2, object 3, object 4... object 10.

[0046] In summary, by filtering products based on behavioral information and user needs, the selected recommended products better meet the personalized needs of users.

[0047] Step S204: Send information consultation requests to the object providers corresponding to the plurality of objects to be recommended, and select at least one recommended object from the plurality of objects to be recommended based on the consultation results.

[0048] Specifically, the object provider refers to the source that can provide detailed information about the object. For example, when the object is a product, the object provider could be a merchant. When the object is a specific service, the object provider could be a service provider; when the object is home design, the object provider could be a designer, and so on. An information inquiry request is a request initiated to the "object provider" to obtain relevant information about the object to be recommended. The inquiry result refers to the information provided by the object provider in response to the information inquiry request. The recommended object refers to the object obtained after further filtering from the objects to be recommended based on the inquiry result.

[0049] Based on this, the intelligent processing unit can establish communication links with the object providers of each object in the object library. For example, when the object is a product, the intelligent processing unit can establish a communication link with the merchant corresponding to the product, and so on. After identifying multiple objects to be recommended, the intelligent processing unit sends information consultation requests to the object providers corresponding to each object. The information consultation requests can be generated by identifying information that cannot be obtained from the object's query information, and then sending the information consultation request to the object provider based on this information to obtain the consultation result. Alternatively, a consultation information library can be pre-set, and the consultation information included in the consultation information library can be sent to the object provider. Or, the consultation information can also be sent by the target account, with the user setting personalized consultation information. After determining the consultation information, a consultation request is generated based on the consultation information and sent to the object provider corresponding to each object to be recommended. Then, the consultation result is obtained, and at least one recommended object is selected from the multiple objects to be recommended based on the consultation result.

[0050] Furthermore, the example of a target object provider sending an information inquiry request to any one of multiple target objects to be recommended will be used for illustration. In this embodiment, the specific implementation method is as follows: In the target object information presentation interface of the target recommended object, the target object attribute information is parsed and obtained; the target object attribute information is matched with the object query information, and it is determined whether there is any missing query information that does not meet the matching conditions in the object query information; if the missing query information exists, consultation content is generated based on the missing query information and an information consultation request carrying the consultation content is sent to the target object provider.

[0051] Specifically, the target object presentation interface refers to the page that displays detailed information about the object, such as the product details page on an e-commerce platform. Target object attribute information refers to the set of various attributes about the object extracted from the "target object information presentation interface" through parsing. Missing query information refers to information in the object query information that cannot be obtained from the target object presentation interface. Consultation content refers to information generated based on the "missing query information" and sent to the "object provider," which can be a natural language question about the missing query information.

[0052] Based on this, the intelligent processing unit can obtain the target object information presentation interface of the recommended object, then extract and identify the content in the target object information presentation interface to obtain information related to the target object, thereby obtaining the target object attribute information. Then, it matches the target attribute information with the object query information to determine if there is any missing query information that does not meet the matching conditions, that is, whether there is any missing query information that the user is interested in but is not displayed in the target object information presentation interface. If so, it can generate consultation content based on the missing query information and send an information consultation request carrying this consultation content to the target object provider.

[0053] Using the previous example, the supplementary query information is an invoice, and the query content generated based on the supplementary query information is "Can you issue an invoice?"

[0054] In summary, by further inquiring about information not included in the object information presentation interface, we can obtain a more comprehensive range of information about the object, thereby providing a more accurate basis for recommendations and making the recommended objects more in line with the user's needs.

[0055] Furthermore, at least one recommended object can be selected from multiple recommended objects based on the sub-consultation results corresponding to each recommended object. In this embodiment, the specific implementation method is as follows: Obtain the sub-consultation results corresponding to the multiple objects to be recommended, and identify the corresponding consultation attribute information for each sub-consultation result; match the consultation attribute information with the object query information, and filter out at least one recommended object that meets the matching conditions from the multiple objects to be recommended.

[0056] Specifically, a sub-consultation result refers to the information consultation request sent to the corresponding provider for each target object and the feedback information received. A sub-consultation result can be a dialogue. Consultation attribute information refers to the structured information extracted after parsing and identifying the sub-consultation results.

[0057] Based on this, after the intelligent processing unit sends an information consultation request to the object provider, it obtains the sub-consultation results corresponding to each recommended object. Then, it identifies the consultation attribute information from the sub-consultation results. Finally, it matches the consultation attribute information with the object query information to filter out recommended objects that match the object query information.

[0058] Continuing with the previous example, send the message "Can you issue an invoice?" to the providers of objects 1, 2, 3, 4...10 respectively. Objects 1, 2, and 3 will reply that invoices can be issued, while the others will reply that invoices cannot be issued. Recommended objects are: Objects 1, 2, and 3.

[0059] In summary, by filtering recommended candidates from those to be recommended based on consultation results, the recommended candidates are made to better meet the user's needs.

[0060] Step S206: Based on the object query information, the consultation results, and the object details information of the at least one recommended object, generate an object recommendation report.

[0061] Specifically, object details refer to the information used to describe the recommended object. An object recommendation report is a report that presents the recommended object, the basis for the recommendation, and the recommendation process to the user. The object recommendation report displays the recommended object and the basis for the recommendation. It can be a static report or an editable report. Users can select objects of interest in the object recommendation report to take further action. For example, when the object recommendation report is about product recommendations, users can select target objects and the quantity of target objects in the report, and then purchase them.

[0062] Based on this, after obtaining object query information, consultation results, and object detail information, the intelligent processing unit can process the above information to obtain the final object recommendation report. Specifically, the intelligent processing unit can summarize the object query information to obtain the object query information section of the object recommendation report, which can be done using a large language model. It can also parse and summarize the consultation results to obtain the consultation information section of the object recommendation report. Specifically, it can obtain the sub-consultation results corresponding to each object in the consultation results, and then perform semantic analysis and summarization on the sub-consultation results. A large language model can be used to extract key information from the sub-consultation results to obtain the corresponding summary. The object recommendation report can include summaries of sub-consultation results corresponding to each object to be recommended, or it can only include summaries of sub-consultation results corresponding to the recommended object. Object detail information can be parsed and obtained from the object information presentation interface. Specifically, it can extract, parse, and integrate various information about objects included on the object information presentation page to obtain the corresponding object detail information.

[0063] Furthermore, an object recommendation report can be generated based on a summary of the object query information, the sub-consultation content of each recommended object, and the object details of each recommended object. In this embodiment, the specific implementation is as follows: Based on the object query information, structured object query content is generated; in the object information presentation interface corresponding to each of the at least one recommended object, object attribute information and object links are parsed and obtained, and object detail information is generated based on the object attribute information and object links; sub-consultation content corresponding to each of the at least one recommended object is generated based on the consultation result; and an object recommendation report is generated based on the object query content, the object detail information corresponding to each of the at least one recommended object, and the sub-consultation content corresponding to each of the at least one recommended object.

[0064] Specifically, object query content refers to the summary generated after parsing and extracting the original "object query information". Object information presentation page refers to the page displaying detailed information about the object, such as the product details page on an e-commerce platform. Object attribute information refers to the set of various attributes about the object extracted from the "object information presentation interface" through parsing. Object link refers to a page (URL) that allows access to the recommended object; for example, an object link could be a link to the object purchase page or a link to the object information presentation page. The specific settings for object links can be determined by relevant technical personnel, and this embodiment does not impose any limitations. Sub-consultation content refers to the summary generated after parsing and extracting the sub-consultation results.

[0065] Based on this, the intelligent processing unit summarizes the object query information to obtain the object query information section of the object recommendation report, which can be summarized using a large language model. It then obtains the sub-consultation results of the recommended objects from the consultation results, summarizes them, and obtains the corresponding sub-consultation content. In the object information presentation interface, it parses and obtains the object attribute information and corresponding object links, and generates object detail information for the recommended objects based on the object attribute information and object links. Finally, it integrates the object query content, the object detail information of each recommended object, and the sub-consultation content of each recommended object to generate the object recommendation report.

[0066] In summary, the object recommendation report includes object query information and various related information about the recommended objects, thus providing a comprehensive basis for users' decision-making.

[0067] Furthermore, evaluation information of the recommended objects can be obtained and added to the object report. In this embodiment, the specific implementation is as follows: Obtain the evaluation information corresponding to each of the at least one recommended object; generate an object recommendation report based on the evaluation information corresponding to each of the at least one recommended object, the object query information, the consultation result, and the object details information of the at least one recommended object.

[0068] Specifically, evaluation information refers to evaluations of an object from a third party (not the target account or the object provider). Evaluation information can include various information such as historical text reviews and ratings. When the object is a product, evaluation information can include user comments on the product and the merchant, and user ratings of the product or the merchant.

[0069] Based on this, the intelligent processing unit can obtain the evaluation information of the recommended objects, analyze the evaluation information to obtain a quality overview of the recommended objects, and add this quality overview to the object recommendation report. For example, it can extract the comments of the recommended objects, determine the ratio of the number of positive comments to the number of negative comments, and generate "more positive comments" when the ratio is greater than a threshold, and "more negative comments" when the ratio is less than the threshold. Evaluation information can also be generated based on object or object provider ratings, etc. The specific method of generating evaluation information is determined by relevant technical personnel, and this embodiment does not impose any limitations. Then, based on the evaluation information corresponding to at least one recommended object, object query information, consultation results, and object details information of at least one recommended object, an object recommendation report is generated.

[0070] In summary, adding evaluation information to the object recommendation report allows users to obtain more comprehensive information about the recommended objects, thereby providing users with richer evidence for decision-making.

[0071] Furthermore, the recommended objects can be sorted based on the evaluation information and the matching degree between the recommended objects and the object query information, and then displayed according to the sorted recommended objects. In this embodiment, the specific implementation method is as follows: A matching score is obtained based on the matching degree between the recommended object information and the object query information of each recommended object, wherein the recommended object information includes object details and sub-consultation results; an evaluation score is obtained based on the evaluation information of each recommended object; a comprehensive score is obtained for each recommended object based on the matching score and the evaluation score; the at least one recommended object is sorted according to the comprehensive score to obtain a recommended object ranking result; an object recommendation report is generated based on the object query information, the recommended objects in the recommended object ranking result, the evaluation information corresponding to the recommended objects, and the recommended object information.

[0072] Specifically, the matching score is a quantitative score used to measure the degree of fit between the complete information of the "recommended object" (object details and sub-consultation results) and the "object query information." The evaluation score is a quantitative score based on third-party evaluations regarding the quality of the object. The comprehensive score is the score obtained after calculating the "matching score" and the "evaluation score."

[0073] Based on this, the intelligent processing unit integrates the object details information and the object attribute information corresponding to the sub-consultation results to obtain the recommended object information for each recommended object. Then, for each recommended object, a matching score is obtained based on the matching degree between the recommended object information and the object query information. Specifically, the matching scores of each attribute of the recommended object can be calculated, and weights can be assigned to each attribute for a weighted calculation to obtain the corresponding matching score for that recommended object. For each recommended object, an evaluation score is obtained based on the evaluation information. Specifically, when the evaluation information is an object rating and an object provider rating, a weighted calculation can be performed to obtain the evaluation score. Furthermore, when the evaluation information is a text evaluation, an evaluation score can be generated based on the ratio of "positive" to "negative" reviews. Then, the matching score and evaluation score are weighted and calculated to obtain the comprehensive score for each recommended object. Finally, the recommended objects are displayed in descending order of comprehensive score, that is, the object details information of the recommended objects and the sub-consultation content summarized from the sub-consultation results are displayed.

[0074] Using the previous example, assuming the overall scores are in descending order as: Object 1, Object 2, and Object 3, the order of the recommended object sections in the object recommendation report is: Object details and sub-consultation content for Object 1, Object details and sub-consultation content for Object 2, and Object details and sub-consultation content for Object 3.

[0075] In summary, displaying recommended items based on comprehensive scores makes the item recommendation report clearer and more intuitive, improves the user's decision-making speed, and thus greatly enhances the user experience.

[0076] The object recommendation method provided in this embodiment can be applied to an intelligent processing unit. The intelligent processing unit obtains object query information sent by a target account and identifies multiple objects to be recommended in the object database that are associated with the object query information. It then sends information consultation requests to the object providers corresponding to each of the multiple objects to be recommended, greatly enriching the information dimensions of the objects. Based on the consultation results, at least one recommended object is selected from the multiple objects to be recommended, making the recommended objects more accurately meet the user's needs. Finally, based on the object query information, consultation results, and detailed information of the object corresponding to at least one recommendation, an object recommendation report is generated, making the recommendation results transparent and well-founded. Using the object recommendation method provided in this embodiment, users only need to provide their needs, without having to search through the object database one by one to obtain an object recommendation report. Further decisions are then made based on the object recommendation report, greatly improving the efficiency of users selecting objects from the object database and thus enhancing the user experience.

[0077] See Figure 3 , Figure 3 The flowchart illustrates another object recommendation method provided according to an embodiment of this specification, applied to a cloud-side device, and specifically includes the following steps.

[0078] Step S302: Obtain object query information sent by the terminal device, and determine multiple objects to be recommended in the object database that are associated with the object query information.

[0079] Furthermore, object query information can be obtained based on object requirement information. In this embodiment, the specific implementation is as follows: Receive object requirement information sent by the target account; parse the object requirement information to obtain the object query information corresponding to the object requirement information.

[0080] Furthermore, if the object requirement information is incomplete, it can be supplemented through prompts. In this embodiment, the specific implementation method is as follows: Receive initial requirement information sent by the target account, compare the initial requirement information with the query template; if the comparison result shows that there is missing requirement information, generate a prompt message based on the missing requirement information and send it to the target account to obtain supplementary requirement information; merge the initial requirement information and the supplementary requirement information to obtain the object requirement information.

[0081] Furthermore, behavioral information of the target account can be obtained, and the target query information can be determined based on the behavioral information and the target's needs. In this embodiment, the specific implementation is as follows: Obtain the behavioral information of the target account; parse the object requirement information and the behavioral information to obtain the object query information corresponding to the object requirement information.

[0082] Furthermore, multiple objects to be recommended can be identified from the object library based on object demand information and behavioral information. In this embodiment, the specific implementation is as follows: Based on the target account's object requirement information, multiple candidate objects matching the object requirement information are filtered out from the object library; the matching degree scores of the multiple candidate objects with the target account's behavioral information are calculated, and the multiple candidate objects are sorted according to the matching degree scores; a set number of candidate objects are selected from the sorted multiple candidate objects as multiple objects to be recommended.

[0083] Step S304: Send information consultation requests to the object providers corresponding to the plurality of objects to be recommended, and select at least one recommended object from the plurality of objects to be recommended based on the consultation results.

[0084] Furthermore, the example of a target object provider sending an information inquiry request to any one of multiple target objects to be recommended will be used for illustration. In this embodiment, the specific implementation method is as follows: In the target object information presentation interface of the target recommended object, the target object attribute information is parsed and obtained; the target object attribute information is matched with the object query information, and it is determined whether there is any missing query information that does not meet the matching conditions in the object query information; if the missing query information exists, consultation content is generated based on the missing query information and an information consultation request carrying the consultation content is sent to the target object provider.

[0085] Furthermore, at least one recommended object can be selected from multiple recommended objects based on the sub-consultation results corresponding to each recommended object. In this embodiment, the specific implementation method is as follows: Obtain the sub-consultation results corresponding to the multiple objects to be recommended, and identify the corresponding consultation attribute information for each sub-consultation result; match the consultation attribute information with the object query information, and filter out at least one recommended object that meets the matching conditions from the multiple objects to be recommended.

[0086] Step S306: Based on the object query information, the consultation result, and the object details information of the at least one recommended object, generate an object recommendation report and send it to the end device.

[0087] Furthermore, an object recommendation report can be generated based on a summary of the object query information, the sub-consultation content of each recommended object, and the object details of each recommended object. In this embodiment, the specific implementation is as follows: Based on the object query information, structured object query content is generated; in the object information presentation interface corresponding to each of the at least one recommended object, object attribute information and object links are parsed and obtained, and object detail information is generated based on the object attribute information and object links; sub-consultation content corresponding to each of the at least one recommended object is generated based on the consultation result; and an object recommendation report is generated based on the object query content, the object detail information corresponding to each of the at least one recommended object, and the sub-consultation content corresponding to each of the at least one recommended object.

[0088] Furthermore, evaluation information of the recommended objects can be obtained and added to the object report. In this embodiment, the specific implementation is as follows: Obtain the evaluation information corresponding to each of the at least one recommended object; generate an object recommendation report based on the evaluation information corresponding to each of the at least one recommended object, the object query information, the consultation result, and the object details information of the at least one recommended object.

[0089] Furthermore, the recommended objects can be sorted based on the evaluation information and the matching degree between the recommended objects and the object query information, and then displayed according to the sorted recommended objects. In this embodiment, the specific implementation method is as follows: A matching score is obtained based on the matching degree between the recommended object information and the object query information of each recommended object, wherein the recommended object information includes object details and sub-consultation results; an evaluation score is obtained based on the evaluation information of each recommended object; a comprehensive score is obtained for each recommended object based on the matching score and the evaluation score; the at least one recommended object is sorted according to the comprehensive score to obtain a recommended object ranking result; an object recommendation report is generated based on the object query information, the recommended objects in the recommended object ranking result, the evaluation information corresponding to the recommended objects, and the recommended object information.

[0090] The description of the object recommendation method in this embodiment can be found in the same or corresponding descriptions in the above embodiments, and will not be repeated here.

[0091] In summary, the object recommendation process is completed on the cloud-side device, which generates an object recommendation report and sends it back to the edge device for use. This eliminates the need to consume processing resources on the edge device, reducing the user's hardware costs by eliminating the need to deploy high-performance processors on the edge device.

[0092] The object recommendation method provided in this embodiment obtains object query information sent by a target account, identifies multiple objects to be recommended in the object database associated with the object query information, and sends information consultation requests to the object providers corresponding to each of the multiple objects to be recommended, greatly enriching the information dimensions of the objects. Then, based on the consultation results, at least one recommended object is selected from the multiple objects to be recommended, making the recommended objects more accurately meet the user's needs. Finally, based on the object query information, consultation results, and detailed information of the object corresponding to at least one recommendation, an object recommendation report is generated, making the recommendation results transparent and well-founded. Using the object recommendation method provided in this embodiment, users only need to provide their needs, without having to search through the object database one by one to obtain an object recommendation report, and then make further decisions based on the object recommendation report, greatly improving the efficiency of users selecting objects in the object database, thereby enhancing the user experience.

[0093] The following is in conjunction with the appendix Figure 4 Taking the application of the object recommendation method provided in this specification in a product recommendation scenario as an example, the object recommendation method will be further explained. Among other things, Figure 4 The flowchart of an object recommendation method provided in one embodiment of this specification is shown, which specifically includes the following steps.

[0094] Step S402: Receive the initial demand information sent by the target account and compare the initial demand information with the query template.

[0095] Step S404: If the comparison result indicates that there is missing requirement information, a prompt message is generated based on the missing requirement information and sent to the target account to obtain supplementary requirement information.

[0096] Step S406: The initial demand information and the supplementary demand information are merged to obtain product demand information.

[0097] Step S408: Obtain the behavior information of the target account, parse the product demand information and the behavior information, and obtain the product query information corresponding to the product demand information.

[0098] Step S410: Based on the product demand information of the target account, select multiple candidate products in the product library that match the product demand information.

[0099] Step S412: Calculate the matching score between the multiple candidate products and the behavioral information of the target account, sort the multiple candidate products according to the matching score, and select a set number of candidate products from the sorted multiple candidate products as multiple products to be recommended.

[0100] Step S414: Send information consultation requests to the product providers corresponding to the plurality of products to be recommended, and select at least one recommended product from the plurality of products to be recommended based on the consultation results.

[0101] In the target product information presentation interface of the target recommended product, the target product attribute information is parsed and obtained; the target product attribute information is matched with the product query information, and it is determined whether there is any missing query information that does not meet the matching conditions in the product query information; if there is any missing query information, based on the missing query information, consultation content is generated and an information consultation request carrying the consultation content is sent to the target product provider.

[0102] Obtain the sub-consultation results corresponding to the multiple products to be recommended, and identify the corresponding consultation attribute information for each sub-consultation result; match the consultation attribute information with the product query information, and filter out at least one recommended product that meets the matching conditions from the multiple products to be recommended.

[0103] Step S416: Based on the product query information, generate structured product query content; in the product information presentation interface corresponding to each of the at least one recommended product, parse and obtain product attribute information and product links, and generate product detail information based on the product attribute information and product links.

[0104] Step S418: Generate sub-consultation content corresponding to each of the at least one recommended product based on the consultation results.

[0105] Step S420: Based on the matching degree between the recommended product information and the product query information of each recommended product, a matching degree score is obtained.

[0106] Specifically, the recommended product information includes product details and sub-consultation results.

[0107] Step S422: Obtain the evaluation information corresponding to the at least one recommended product, obtain the evaluation score based on the evaluation information of each recommended product, and obtain the comprehensive score corresponding to each recommended product based on the matching score and the evaluation score.

[0108] Step S424: Sort the at least one recommended product according to the comprehensive score to obtain the recommended product ranking result.

[0109] Step S426: Generate a product recommendation report based on the product query information, the recommended products in the recommended product ranking results, the evaluation information corresponding to the recommended products, and the recommended product information.

[0110] Specifically, the behavioral information of the target account can include product orders corresponding to the target account, browsing history of the target account, merchant favorites and product favorites of the target account, etc.

[0111] For example, such as Figure 5 The image shown is a schematic diagram of a product recommendation report. Assume a company's purchasing agent needs to buy 100 black ballpoint pens with a total budget of 5200 yuan. The product search terms are: black ink, minimum budget of 5 yuan per pen, smooth writing, quick-drying and waterproof, suitable for daily office writing, and an invoice required. The generated product recommendation report would look like this: Figure 5 As shown.

[0112] In summary, the object recommendation method provided in this embodiment can be applied to an intelligent processing unit. The intelligent processing unit obtains product query information sent by a target account and identifies multiple products in the product database that are associated with the query information. It then sends information inquiry requests to the product providers corresponding to each of these products, greatly enriching the product information dimensions. Based on the inquiry results, at least one recommended product is selected from the multiple products to be recommended, making the recommended products more accurately meet the user's needs. Finally, based on the product query information, the inquiry results, and the product details corresponding to at least one recommendation, a product recommendation report is generated, making the recommendation results transparent and well-founded. Using the product recommendation method provided in this embodiment, users only need to provide their needs without having to search through the product database one by one to obtain a product recommendation report. Further decisions are then made based on the product recommendation report, greatly improving the efficiency of users selecting products from the product database and thus enhancing the user experience.

[0113] Corresponding to the above method embodiments, this specification also provides embodiments of an object recommendation device. Figure 6 A schematic diagram of an object recommendation device according to one embodiment of this specification is shown. Figure 6 As shown, this device can be applied to an intelligent processing unit, and the device includes: The first acquisition module 602 is configured to acquire object query information sent by the target account and determine multiple objects to be recommended in the object library that are associated with the object query information; The first sending module 604 is configured to send information consultation requests to the object providers corresponding to the plurality of objects to be recommended, and to select at least one recommended object from the plurality of objects to be recommended based on the consultation results. The first generation module 606 is configured to generate an object recommendation report based on the object query information, the consultation results, and the object details information of the at least one recommended object.

[0114] In an optional embodiment, obtaining the object query information sent by the target account includes: Receive object request information sent by the target account; Parse the object requirement information to obtain the object query information corresponding to the object requirement information.

[0115] In an optional embodiment, receiving the object request information sent by the target account includes: Receive initial request information sent by the target account and compare the initial request information with the query template; If the comparison result indicates that there is missing requirement information, a prompt message is generated based on the missing requirement information and sent to the target account to obtain supplementary requirement information; The initial requirement information and the supplementary requirement information are merged to obtain the object requirement information.

[0116] In an optional embodiment, it further includes: Obtain behavioral information of the target account; The step of parsing the object requirement information and obtaining the object query information corresponding to the object requirement information includes: Parse the object requirement information and the behavior information to obtain the object query information corresponding to the object requirement information.

[0117] In an optional embodiment, determining multiple objects to be recommended in the object database that are associated with the object query information includes: Based on the target account's object requirement information, multiple candidate objects matching the object requirement information are filtered out from the object library; Calculate the matching score between the multiple candidate objects and the behavioral information of the target account, and sort the multiple candidate objects according to the matching score; A set number of candidate objects are selected from the sorted candidate objects as multiple objects to be recommended.

[0118] In an optional embodiment, sending an information inquiry request to the target object provider corresponding to any one of the plurality of target objects to be recommended includes: In the target object information presentation interface of the target recommended object, the target object attribute information is parsed and obtained; The target object attribute information and the object query information are matched, and it is determined whether there is any missing query information in the object query information that does not meet the matching conditions. In the event of missing query information, based on the missing query information, a consultation content is generated and an information consultation request carrying the consultation content is sent to the target object provider.

[0119] In an optional embodiment, the step of selecting at least one recommended object from the plurality of objects to be recommended based on the consultation results includes: Obtain the sub-consultation results corresponding to the multiple objects to be recommended, and identify the corresponding consultation attribute information for each sub-consultation result; The consultation attribute information is matched with the object query information, and at least one recommended object that meets the matching conditions is selected from the plurality of objects to be recommended.

[0120] In an optional embodiment, generating an object recommendation report based on the object query information, the consultation results, and the object details information of the at least one recommended object includes: Based on the object query information, structured object query content is generated; In the object information presentation interface corresponding to each of the at least one recommended object, object attribute information and object links are parsed and obtained, and object detail information is generated based on the object attribute information and object links. Based on the consultation results, generate sub-consultation content corresponding to each of the at least one recommended object; Based on the object query content, the object details information corresponding to the at least one recommended object, and the sub-consultation content corresponding to the at least one recommended object, an object recommendation report is generated.

[0121] In an optional embodiment, it further includes: Obtain the evaluation information corresponding to each of the at least one recommended object; The process of generating an object recommendation report based on the object query information, the consultation results, and the object details information of at least one recommended object includes: An object recommendation report is generated based on the evaluation information corresponding to the at least one recommended object, the object query information, the consultation results, and the object details information of the at least one recommended object.

[0122] In an optional embodiment, generating an object recommendation report based on the evaluation information corresponding to the at least one recommended object, the object query information, the consultation result, and the object details information of the at least one recommended object includes: A matching score is obtained based on the matching degree between the recommended object information and the object query information of each recommended object, wherein the recommended object information includes object details information and sub-consultation results; The evaluation score is obtained based on the evaluation information of each recommended object; Based on the matching score and the evaluation score, a comprehensive score is obtained for each recommended object; The at least one recommended object is sorted according to its comprehensive score to obtain the ranking result of the recommended object; Based on the object query information, the recommended objects in the recommended object ranking results, the evaluation information corresponding to the recommended objects, and the recommended object information, an object recommendation report is generated.

[0123] The object recommendation device provided in this embodiment can be applied to an intelligent processing unit to obtain object query information sent by a target account, identify multiple objects to be recommended in the object library associated with the object query information, and send information consultation requests to the object providers corresponding to each of the multiple objects to be recommended, greatly enriching the information dimensions of the objects. Then, based on the consultation results, at least one recommended object is selected from the multiple objects to be recommended, making the recommended object more accurately meet the user's needs. Finally, based on the object query information, consultation results, and detailed information of the object corresponding to at least one recommendation, an object recommendation report is generated, making the recommendation results transparent and well-founded. Using the object recommendation device provided in this embodiment, users only need to provide their needs, without having to search through the object library one by one to obtain an object recommendation report, and then make further decisions based on the object recommendation report, greatly improving the efficiency of users selecting objects in the object library, thereby enhancing the user experience.

[0124] The above is an illustrative scheme of an object recommendation device according to this embodiment. It should be noted that the technical solution of this object recommendation device and the technical solution of the object recommendation method described above belong to the same concept. For details not described in detail in the technical solution of the object recommendation device, please refer to the description of the technical solution of the object recommendation method described above.

[0125] Corresponding to the above method embodiments, this specification also provides another object recommendation device embodiment, applied to cloud-side equipment. Figure 7 A schematic diagram of another object recommendation device provided in one embodiment of this specification is shown. Figure 7 As shown, the device includes: The second acquisition module 702 is configured to acquire object query information sent by the end device and determine multiple objects to be recommended in the object library that are associated with the object query information; The second sending module 704 is configured to send information consultation requests to the object providing end corresponding to the plurality of objects to be recommended, and to select at least one recommended object from the plurality of objects to be recommended based on the consultation results. The second generation module 706 is configured to generate an object recommendation report and send it to the end device based on the object query information, the consultation results, and the object details information of the at least one recommended object.

[0126] In an optional embodiment, obtaining the object query information sent by the end-side device includes: The receiving end device sends object requirement information; Parse the object requirement information to obtain the object query information corresponding to the object requirement information.

[0127] In an optional embodiment, the object requirement information sent by the receiving end device includes: The receiving end-side device sends initial demand information, and compares the initial demand information with the query template; If the comparison result indicates that there is missing requirement information, a prompt message is generated based on the missing requirement information and sent to the end device to obtain supplementary requirement information. The initial requirement information and the supplementary requirement information are merged to obtain the object requirement information.

[0128] In an optional embodiment, it further includes: Obtain the behavior information of the terminal device; The step of parsing the object requirement information and obtaining the object query information corresponding to the object requirement information includes: Parse the object requirement information and the behavior information to obtain the object query information corresponding to the object requirement information.

[0129] In an optional embodiment, determining multiple objects to be recommended in the object database that are associated with the object query information includes: Based on the object requirement information of the terminal device, multiple candidate objects that match the object requirement information are selected from the object library; Calculate the matching score between the plurality of candidate objects and the behavioral information of the terminal device, and sort the plurality of candidate objects according to the matching score; A set number of candidate objects are selected from the sorted candidate objects as multiple objects to be recommended.

[0130] In an optional embodiment, sending an information inquiry request to the target object provider corresponding to any one of the plurality of target objects to be recommended includes: In the target object information presentation interface of the target recommended object, the target object attribute information is parsed and obtained; The target object attribute information and the object query information are matched, and it is determined whether there is any missing query information in the object query information that does not meet the matching conditions. In the event of missing query information, based on the missing query information, a consultation content is generated and an information consultation request carrying the consultation content is sent to the target object provider.

[0131] In an optional embodiment, the step of selecting at least one recommended object from the plurality of objects to be recommended based on the consultation results includes: Obtain the sub-consultation results corresponding to the multiple objects to be recommended, and identify the corresponding consultation attribute information for each sub-consultation result; The consultation attribute information is matched with the object query information, and at least one recommended object that meets the matching conditions is selected from the plurality of objects to be recommended.

[0132] In an optional embodiment, generating an object recommendation report and sending it to the edge device based on the object query information, the consultation results, and the object details information of the at least one recommended object includes: Based on the object query information, structured object query content is generated; In the object information presentation interface corresponding to each of the at least one recommended object, object attribute information and object links are parsed and obtained, and object detail information is generated based on the object attribute information and object links. Based on the consultation results, generate sub-consultation content corresponding to each of the at least one recommended object; Based on the object query content, the object details information corresponding to the at least one recommended object, and the sub-consultation content corresponding to the at least one recommended object, an object recommendation report is generated and sent to the end device.

[0133] In an optional embodiment, it further includes: Obtain the evaluation information corresponding to each of the at least one recommended object; The step of generating an object recommendation report and sending it to the edge device based on the object query information, the consultation results, and the object details information of at least one recommended object includes: Based on the evaluation information corresponding to the at least one recommended object, the object query information, the consultation result, and the object details information of the at least one recommended object, an object recommendation report is generated and sent to the end device.

[0134] In an optional embodiment, generating an object recommendation report and sending it to the edge device based on the evaluation information corresponding to the at least one recommended object, the object query information, the consultation result, and the object details information of the at least one recommended object includes: A matching score is obtained based on the matching degree between the recommended object information and the object query information of each recommended object, wherein the recommended object information includes object details information and sub-consultation results; The evaluation score is obtained based on the evaluation information of each recommended object; Based on the matching score and the evaluation score, a comprehensive score is obtained for each recommended object; The at least one recommended object is sorted according to its comprehensive score to obtain the ranking result of the recommended object; Based on the object query information, the recommended objects in the recommended object ranking results, the evaluation information corresponding to the recommended objects, and the recommended object information, an object recommendation report is generated and sent to the end device.

[0135] In summary, the object recommendation process is completed on the cloud-side device, which generates an object recommendation report and sends it back to the edge device for use. This eliminates the need to consume processing resources on the edge device, reducing the user's hardware costs by eliminating the need to deploy high-performance processors on the edge device.

[0136] The object recommendation device provided in this embodiment obtains object query information sent by a target account, identifies multiple objects to be recommended in the object database associated with the object query information, and sends information consultation requests to the object providers corresponding to each of the multiple objects to be recommended, greatly enriching the information dimensions of the objects. Then, based on the consultation results, at least one recommended object is selected from the multiple objects to be recommended, making the recommended objects more accurately meet the user's needs. Finally, based on the object query information, consultation results, and detailed information of the object corresponding to at least one recommendation, an object recommendation report is generated, making the recommendation results transparent and well-founded. Using the object recommendation device provided in this embodiment, users only need to provide their needs, without having to search through the object database one by one to obtain an object recommendation report, and then make further decisions based on the object recommendation report, greatly improving the efficiency of users selecting objects in the object database, thereby enhancing the user experience.

[0137] The above is an illustrative scheme of an object recommendation device according to this embodiment. It should be noted that the technical solution of this object recommendation device and the technical solution of the object recommendation method described above belong to the same concept. For details not described in detail in the technical solution of the object recommendation device, please refer to the description of the technical solution of the object recommendation method described above.

[0138] Figure 8 A structural block diagram of a computing device 800 according to one embodiment of this specification is shown. The components of the computing device 800 include, but are not limited to, a memory 810 and a processor 820. The processor 820 is connected to the memory 810 via a bus 830, and a database 850 is used to store data.

[0139] The computing device 800 also includes an access device 840, which enables the computing device 800 to communicate via one or more networks 860. Examples of these networks include Public Switched Telephone Network (PSTN), Local Area Network (LAN), Wide Area Network (WAN), Personal Area Network (PAN), or combinations of communication networks such as the Internet. The access device 840 may include one or more of any type of wired or wireless network interface (e.g., a network interface card (NIC)), such as an IEEE 802.11 Wireless Local Area Network (WLAN) wireless interface, a Wi-MAX (Worldwide Interoperability for Microwave Access) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, or a Near Field Communication (NFC) interface.

[0140] In one embodiment of this specification, the above-described components of the computing device 800 and Figure 8 Other components, not shown, can also be connected to each other, for example, via a bus. It should be understood that... Figure 8 The block diagram of the computing device shown is for illustrative purposes only and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0141] The computing device 800 can be any type of stationary or mobile computing device, including mobile computers or mobile computing devices (e.g., tablet computers, personal digital assistants, laptop computers, notebook computers, netbooks, etc.), mobile phones (e.g., smartphones), wearable computing devices (e.g., smartwatches, smart glasses, etc.) or other types of mobile devices, or stationary computing devices such as desktop computers or personal computers (PCs). The computing device 800 can also be a mobile or stationary server.

[0142] The processor 820 is configured to execute the following computer-executable instructions, which, when executed by the processor, implement the steps of the above-described object recommendation method.

[0143] The above is an illustrative scheme of a computing device according to this embodiment. It should be noted that the technical solution of this computing device and the technical solution of the object recommendation method described above belong to the same concept. For details not described in detail in the technical solution of the computing device, please refer to the description of the technical solution of the object recommendation method described above.

[0144] An embodiment of this specification also provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the object recommendation method described above.

[0145] The above is an illustrative scheme of a computer-readable storage medium according to this embodiment. It should be noted that the technical solution of this storage medium belongs to the same concept as the technical solution of the object recommendation method described above. Details not described in detail in the technical solution of the storage medium can be found in the description of the technical solution of the object recommendation method described above.

[0146] An embodiment of this specification also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the steps of the object recommendation method described above.

[0147] The above is an illustrative scheme of a computer program product according to this embodiment. It should be noted that the technical solution of this computer program product and the technical solution of the object recommendation method described above belong to the same concept. For details not described in detail in the technical solution of the computer program product, please refer to the description of the technical solution of the object recommendation method described above.

[0148] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0149] The computer program / instructions include computer program code, which may be in the form of source code, object code, executable file, or certain intermediate forms. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium may be appropriately added or removed according to the requirements of patent practice. For example, in some regions, according to patent practice, computer-readable media may not include electrical carrier signals and telecommunication signals.

[0150] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments in this specification are not limited to the described order of actions, because according to the embodiments in this specification, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the embodiments in this specification.

[0151] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0152] The preferred embodiments disclosed above are merely illustrative of this specification. Optional embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the embodiments described herein. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the embodiments, thereby enabling those skilled in the art to better understand and utilize this specification. This specification is limited only by the claims and their full scope and equivalents.

Claims

1. An object recommendation method, applied to an intelligent processing unit, comprising: Obtain object query information sent by the target account, and identify multiple objects to be recommended in the object database that are associated with the object query information; Send information consultation requests to the object providers corresponding to the plurality of objects to be recommended, and select at least one recommended object from the plurality of objects to be recommended based on the consultation results; Based on the object query information, the consultation results, and the object details information of the at least one recommended object, an object recommendation report is generated.

2. The object recommendation method according to claim 1, wherein obtaining the object query information sent by the target account includes: Receive object request information sent by the target account; Parse the object requirement information to obtain the object query information corresponding to the object requirement information.

3. The object recommendation method according to claim 2, wherein receiving object demand information sent by the target account includes: Receive initial request information sent by the target account and compare the initial request information with the query template; If the comparison result indicates that there is missing requirement information, a prompt message is generated based on the missing requirement information and sent to the target account to obtain supplementary requirement information; The initial requirement information and the supplementary requirement information are merged to obtain the object requirement information.

4. The object recommendation method according to claim 2 further includes: Obtain behavioral information of the target account; The step of parsing the object requirement information and obtaining the object query information corresponding to the object requirement information includes: Parse the object requirement information and the behavior information to obtain the object query information corresponding to the object requirement information.

5. The object recommendation method according to claim 1, wherein determining multiple objects to be recommended in the object database that are associated with the object query information includes: Based on the target account's object requirement information, multiple candidate objects matching the object requirement information are filtered out from the object library; Calculate the matching score between the multiple candidate objects and the behavioral information of the target account, and sort the multiple candidate objects according to the matching score; A set number of candidate objects are selected from the sorted candidate objects as multiple objects to be recommended.

6. The object recommendation method according to claim 1, comprising sending an information consultation request to the target object provider corresponding to any one of the plurality of objects to be recommended, including: In the target object information presentation interface of the target recommended object, the target object attribute information is parsed and obtained; The target object attribute information and the object query information are matched, and it is determined whether there is any missing query information in the object query information that does not meet the matching conditions. In the event of missing query information, based on the missing query information, a consultation content is generated and an information consultation request carrying the consultation content is sent to the target object provider.

7. The object recommendation method according to claim 1, wherein selecting at least one recommended object from the plurality of objects to be recommended based on the consultation results includes: Obtain the sub-consultation results corresponding to the multiple objects to be recommended, and identify the corresponding consultation attribute information for each sub-consultation result; The consultation attribute information is matched with the object query information, and at least one recommended object that meets the matching conditions is selected from the plurality of objects to be recommended.

8. The object recommendation method according to claim 1, wherein generating an object recommendation report based on the object query information, the consultation result, and the object details information of the at least one recommended object includes: Based on the object query information, structured object query content is generated; In the object information presentation interface corresponding to each of the at least one recommended object, object attribute information and object links are parsed and obtained, and object detail information is generated based on the object attribute information and object links. Based on the consultation results, generate sub-consultation content corresponding to each of the at least one recommended object; Based on the object query content, the object details information corresponding to the at least one recommended object, and the sub-consultation content corresponding to the at least one recommended object, an object recommendation report is generated.

9. The object recommendation method according to claim 1, further comprising: Obtain the evaluation information corresponding to each of the at least one recommended object; The process of generating an object recommendation report based on the object query information, the consultation results, and the object details information of at least one recommended object includes: An object recommendation report is generated based on the evaluation information corresponding to the at least one recommended object, the object query information, the consultation results, and the object details information of the at least one recommended object.

10. The object recommendation method according to claim 9, wherein generating an object recommendation report based on the evaluation information corresponding to the at least one recommended object, the object query information, the consultation result, and the object detail information of the at least one recommended object includes: A matching score is obtained based on the matching degree between the recommended object information and the object query information of each recommended object, wherein the recommended object information includes object details information and sub-consultation results; The evaluation score is obtained based on the evaluation information of each recommended object; Based on the matching score and the evaluation score, a comprehensive score is obtained for each recommended object; The at least one recommended object is sorted according to its comprehensive score to obtain the ranking result of the recommended object; Based on the object query information, the recommended objects in the recommended object ranking results, the evaluation information corresponding to the recommended objects, and the recommended object information, an object recommendation report is generated.

11. An object recommendation method, applied to cloud-side devices, comprising: Obtain object query information sent by the terminal device, and determine multiple objects to be recommended in the object database that are associated with the object query information; Send information consultation requests to the object providers corresponding to the plurality of objects to be recommended, and select at least one recommended object from the plurality of objects to be recommended based on the consultation results; Based on the object query information, the consultation results, and the object details information of the at least one recommended object, an object recommendation report is generated and sent to the end device.

12. An object recommendation device, applied to an intelligent processing unit, comprising: The first acquisition module is configured to acquire object query information sent by the target account and determine multiple objects to be recommended in the object library that are associated with the object query information; The first sending module is configured to send information consultation requests to the object providers corresponding to the plurality of objects to be recommended, and to select at least one recommended object from the plurality of objects to be recommended based on the consultation results. The first generation module is configured to generate an object recommendation report based on the object query information, the consultation results, and the object details information of the at least one recommended object.

13. An object recommendation device, applied to cloud-side equipment, comprising: The second acquisition module is configured to acquire object query information sent by the end device and determine multiple objects to be recommended in the object library that are associated with the object query information; The second sending module is configured to send information consultation requests to the object providers corresponding to the plurality of objects to be recommended, and to select at least one recommended object from the plurality of objects to be recommended based on the consultation results. The second generation module is configured to generate an object recommendation report and send it to the end device based on the object query information, the consultation results, and the object details information of the at least one recommended object.

14. A computing device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the method according to any one of claims 1 to 11.

15. A computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 11.

16. A computer program product comprising a computer program or instructions which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 11.

Images