Methods, apparatus, devices, storage media, and programs for sending display information

By determining the importance score of candidate display objects in the search information and using a rewriting model to generate more accurate rewritten information, the problem of mismatch between user intent and existing technologies is solved, thereby improving the accuracy of display objects and user experience.

CN123115506APending Publication Date: 2026-11-17BUSINESS LINE COMMERCIAL PTE LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611544792.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-30
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing technologies, the initial search information entered by the user cannot accurately match the platform's retrieval system, resulting in the retrieved display objects not matching the user's intent. Existing rewriting methods are crude, reducing the accuracy of rewritten information and displayed objects.

Method used

By acquiring multiple information units from the initial search information, candidate display objects are determined based on semantic matching and keyword matching models. The importance score of each information unit to the candidate display object is calculated, and a rewriting model is used to generate more accurate rewriting information to guide the focus on key information during the rewriting process.

Benefits of technology

It improved the accuracy of displayed objects, enhanced the accuracy of rewritten information, and improved the user search experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123115506A_ABST
    Figure CN123115506A_ABST
Patent Text Reader

Abstract

This disclosure provides a method, apparatus, device, storage medium, and program product for sending display information, belonging to the field of computer technology. The method includes: acquiring initial search information and multiple information units included in the initial search information; determining candidate display objects from multiple objects to be displayed based on the initial search information; determining the importance score of each information unit to the display information of each candidate display object based on the display information of the multiple information units and the multiple candidate display objects; determining rewritten information corresponding to the initial search information based on the initial search information, the importance score corresponding to each information unit, and a rewriting model; and determining the display object from the multiple objects to be displayed based on the rewritten information, and sending the display information of the display object to a target account. Using this disclosure improves the accuracy of the rewritten information, thereby improving the accuracy of the determined display objects.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to a method, apparatus, device, storage medium, and program product for transmitting display information. Background Technology

[0002] Many current websites and applications have search functions. They use the initial search information entered by the user to identify matching items from a stored list of potential displays and send them to the user's account for recall and display. However, because the initial search information often doesn't align with the platform's retrieval system's processing methods, the retrieved items may not match the user's intent. For example, if a user enters "cheap shoes," the platform cannot understand the specific price range for "cheap," thus failing to achieve effective recall. To address this, many platforms rewrite the initial search information to make it clearer and more suitable for the platform's retrieval system.

[0003] The usual rewriting method is to directly input the initial search information into the rewriting model and obtain the output rewritten information.

[0004] However, the above rewriting method is rather crude, which reduces the accuracy of the rewritten information obtained, and thus reduces the accuracy of the display object determined based on the rewritten information. Summary of the Invention

[0005] This disclosure provides a method for sending display information, which can solve the technical problems existing in related technologies. The technical solution is as follows: Firstly, a method for sending display information is provided, the method comprising: Obtain the initial search information input by the target account and the multiple information units included in the initial search information; Based on the initial search information, candidate display objects are determined from the plurality of objects to be displayed; Based on the display information of the multiple information units and multiple candidate display objects, determine the importance score of each information unit to the display information of each candidate display object; Based on the initial search information, the importance score corresponding to each information unit, and the rewriting model, the rewriting information corresponding to the initial search information is determined; Based on the rewritten information, a display object is determined from the plurality of objects to be displayed, and the display information of the display object is sent to the target account.

[0006] In one possible implementation, the importance score includes positive attribution values ​​and / or negative attribution values.

[0007] In one possible implementation, determining candidate display objects from the plurality of objects to be displayed based on the initial search information includes: Based on the initial search information, the plurality of objects to be displayed, and the semantic matching model, a first candidate display object corresponding to the initial search information is determined from the plurality of objects to be displayed. Based on the initial search information, the plurality of objects to be displayed, and the keyword matching model, a second candidate object corresponding to the initial search information is determined from the plurality of objects to be displayed. The candidate display object is determined based on the first candidate display object and the second candidate display object.

[0008] In one possible implementation, determining the first candidate display object corresponding to the initial search information from among the multiple objects to be displayed, based on the initial search information, the multiple objects to be displayed, and the semantic matching model, includes: Based on the initial search information, the multiple objects to be displayed, and the semantic matching model, determine the semantic matching degree between the initial search information and each object to be displayed; Based on the semantic matching degree, the first candidate display object corresponding to the initial search information is determined among the multiple objects to be displayed; The step of determining the second candidate display object corresponding to the initial search information from among the multiple objects to be displayed, based on the initial search information, the multiple objects to be displayed, and the keyword matching model, includes: Based on the initial search information, the multiple objects to be displayed, and the keyword matching model, determine the keyword matching degree between the initial search information and each object to be displayed; Based on the keyword matching degree, a second candidate display object corresponding to the initial search information is determined from the plurality of objects to be displayed; The step of determining the candidate display object based on the first candidate display object and the second candidate display object includes: Based on the semantic matching degree corresponding to the first candidate display object and the keyword matching degree corresponding to the second candidate display object, the candidate display object is determined from the first candidate display object and the second candidate display object.

[0009] In one possible implementation, determining the candidate display object from the first candidate display object and the second candidate display object based on the semantic matching degree corresponding to the first candidate display object and the keyword matching degree corresponding to the second candidate display object includes: The first ranking information of the first candidate display object is determined based on the semantic matching degree; The second ranking information of the second candidate display object is determined based on the keyword matching degree; Based on the first sorting information and the second sorting information, the third sorting information corresponding to each object to be displayed in the set of display objects composed of the first candidate display objects and the second candidate display objects is determined; Based on the third sorting information, the candidate display objects are determined in the set of display objects.

[0010] In one possible implementation, the method further includes: Based on the multiple information units and the multiple attribute information corresponding to the multiple objects to be displayed, the target attribute information corresponding to each information unit is determined from the multiple attribute information; The process of determining the rewrite information corresponding to the initial search information based on the initial search information, the importance score value corresponding to each information unit, and the rewrite model includes: Based on the initial search information, the importance score corresponding to each information unit, the target attribute information corresponding to each information unit, and the rewriting model, the rewriting information corresponding to the initial search information is determined.

[0011] In one possible implementation, the method further includes: For each target attribute information, based on the importance score of the information unit corresponding to the target attribute information, the attribute importance score corresponding to each target attribute information is determined; The process of determining the rewriting information corresponding to the initial search information based on the initial search information, the importance score corresponding to each information unit, the target attribute information corresponding to each information unit, and the rewriting model includes: Based on the initial search information, the importance score corresponding to each information unit, the target attribute information corresponding to each information unit, the attribute importance score corresponding to each target attribute information, and the rewriting model, the rewriting information corresponding to the initial search information is determined.

[0012] In one possible implementation, the method further includes: The rewritten information and / or the displayed object are verified based on preset verification rules. If the rewritten information does not conform to the preset verification rules, the rewritten information is updated, and the display object is determined based on the updated rewritten information.

[0013] Secondly, a device for sending display information is provided, the device comprising: The acquisition module is used to acquire the initial search information input by the target account and the multiple information units included in the initial search information; The first determining module is used to determine candidate display objects from the plurality of objects to be displayed based on the initial search information. The second determining module is used to determine the importance score of each information unit to the display information of each candidate display object based on the display information of the multiple information units and the multiple candidate display objects. The third determining module is used to determine the rewriting information corresponding to the initial search information based on the initial search information, the importance score value corresponding to each information unit, and the rewriting model. The sending module is used to determine the display object from the plurality of objects to be displayed based on the rewritten information, and send the display information of the display object to the target account.

[0014] In one possible implementation, the importance score includes positive attribution values ​​and / or negative attribution values.

[0015] In one possible implementation, the first determining module is configured to: Based on the initial search information, the plurality of objects to be displayed, and the semantic matching model, a first candidate display object corresponding to the initial search information is determined from the plurality of objects to be displayed. Based on the initial search information, the plurality of objects to be displayed, and the keyword matching model, a second candidate object corresponding to the initial search information is determined from the plurality of objects to be displayed. The candidate display object is determined based on the first candidate display object and the second candidate display object.

[0016] In one possible implementation, the first determining module is configured to: Based on the initial search information, the multiple objects to be displayed, and the semantic matching model, determine the semantic matching degree between the initial search information and each object to be displayed; Based on the semantic matching degree, the first candidate display object corresponding to the initial search information is determined among the multiple objects to be displayed; Based on the initial search information, the multiple objects to be displayed, and the keyword matching model, determine the keyword matching degree between the initial search information and each object to be displayed; Based on the keyword matching degree, a second candidate display object corresponding to the initial search information is determined from the plurality of objects to be displayed; Based on the semantic matching degree corresponding to the first candidate display object and the keyword matching degree corresponding to the second candidate display object, the candidate display object is determined from the first candidate display object and the second candidate display object.

[0017] In one possible implementation, the first determining module is configured to: The first ranking information of the first candidate display object is determined based on the semantic matching degree; The second ranking information of the second candidate display object is determined based on the keyword matching degree; Based on the first sorting information and the second sorting information, the third sorting information corresponding to each object to be displayed in the set of display objects composed of the first candidate display objects and the second candidate display objects is determined; Based on the third sorting information, the candidate display objects are determined in the set of display objects.

[0018] In one possible implementation, the second determining module is further configured to: Based on the multiple information units and the multiple attribute information corresponding to the multiple objects to be displayed, the target attribute information corresponding to each information unit is determined from the multiple attribute information; The third determining module is used for: Based on the initial search information, the importance score corresponding to each information unit, the target attribute information corresponding to each information unit, and the rewriting model, the rewriting information corresponding to the initial search information is determined.

[0019] In one possible implementation, the second determining module is further configured to: For each target attribute information, based on the importance score of the information unit corresponding to the target attribute information, the attribute importance score corresponding to each target attribute information is determined; The third determining module is used for: Based on the initial search information, the importance score corresponding to each information unit, the target attribute information corresponding to each information unit, the attribute importance score corresponding to each target attribute information, and the rewriting model, the rewriting information corresponding to the initial search information is determined.

[0020] In one possible implementation, the sending module is further configured to: The rewritten information and / or the displayed object are verified based on preset verification rules. If the rewritten information does not conform to the preset verification rules, the rewritten information is updated, and the display object is determined based on the updated rewritten information.

[0021] Thirdly, a computer device is provided, the computer device including a processor and a memory, the memory storing at least one instruction, the instruction being loaded and executed by the processor to perform the operation of the method for sending display information.

[0022] Fourthly, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, the instruction being loaded and executed by a processor to implement the operation performed by the method of sending display information.

[0023] Fifthly, a computer program product is provided, the computer program product including at least one instruction, the at least one instruction being loaded and executed by a processor to implement the operation performed by the method of sending display information.

[0024] The beneficial effects of the technical solution provided in this disclosure are as follows: The solution mentioned in this disclosure can identify candidate display objects that match the initial search information. These candidate display objects initially reflect the intent expressed by the initial search information. Then, the importance score of each information unit of the initial search information to each candidate display object can be calculated. This importance score reflects the contribution of the word unit to the recall of the candidate display object. Then, based on the initial search information, the importance score of each information unit to each candidate display object, and the rewriting model, the rewritten information corresponding to the initial search information can be obtained. In the above rewriting process, the importance score can guide the rewriting model to identify which information units have a greater impact on accurately recalling display objects, thereby helping the rewriting model focus on the key information to obtain more accurate rewritten information, thereby improving the accuracy of the identified display objects. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a flowchart of a method for sending display information provided in an embodiment of this disclosure; Figure 2 This is a flowchart of a method for determining a display object provided in an embodiment of this disclosure; Figure 3 This is a flowchart of a method for determining candidate display objects provided in an embodiment of this disclosure; Figure 4 This is a flowchart of a method for determining a display object provided in an embodiment of this disclosure; Figure 5 This is a flowchart of a method for determining a display object provided in an embodiment of this disclosure; Figure 6 This is a flowchart of a method for determining a display object provided in an embodiment of this disclosure; Figure 7 This is a schematic diagram of the structure of a device for sending display information according to an embodiment of this disclosure; Figure 8 This is a structural block diagram of a terminal provided in an embodiment of this disclosure; Figure 9 This is a structural block diagram of a server provided in an embodiment of this disclosure. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this disclosure clearer, the embodiments of this disclosure will be described in further detail below with reference to the accompanying drawings.

[0028] Rewriting methods in related technologies typically involve directly inputting initial search information into a rewriting model to obtain output rewritten information. However, this method is rather crude, reducing the accuracy of the obtained rewritten information and consequently decreasing the accuracy of the displayed objects determined based on that information.

[0029] In response, this disclosure provides a method for sending display information, which can be implemented by a computer device. The computer device can be a terminal or a server, etc. The terminal can be a desktop computer, laptop computer, tablet computer, mobile phone, etc. The server can include a single server or a server cluster, etc.

[0030] In the method for sending display information provided in this embodiment, candidate display objects matching initial search information can be identified. These candidate display objects initially reflect the intent expressed by the initial search information. Then, the importance score of each information unit of the initial search information to each candidate display object can be calculated. This importance score reflects the contribution of the word unit to recalling the candidate display object. Then, based on the initial search information, the importance score of each information unit to each candidate display object, and the rewriting model, rewritten information corresponding to the initial search information can be obtained. In the above rewriting process, the importance score can guide the rewriting model to identify which information units have a greater impact on accurately recalling display objects, thereby helping the rewriting model focus on the key information to obtain more accurate rewritten information, thereby improving the accuracy of the identified display objects.

[0031] The method for sending display information according to embodiments of this disclosure will now be described with reference to the accompanying drawings. This method can be implemented by a computer device; see [link to documentation]. Figure 1 and Figure 2The method may include the following processing steps: 101. Obtain the initial search information entered by the target account and the multiple information units included in the initial search information.

[0032] In implementation, users can open the platform on their terminals and access a search page. The search page displays a search box where users can enter their desired search terms and click the search option. This triggers the user terminal to retrieve the content entered in the search box, thus obtaining the initial search information from the target account corresponding to that user.

[0033] In this embodiment of the disclosure, the initial search information can be any reasonable information, or information corresponding to any language. For example, it can be Chinese "red beach wedding chiffon dress priced under 300 yuan" or English "red beach wedding chiffon dress under $45".

[0034] The platform can be any webpage or application, for example, a shopping application. The initial search information can be information for searching for products, and the subsequent objects to be displayed can be various products that can be displayed in the shopping application.

[0035] After obtaining the initial search information, multiple information units included in the initial search information can be obtained according to preset rules. The information unit can be any information unit that makes up the initial search information, such as multiple word segments or multiple word elements included in the initial search information, etc. This disclosure embodiment does not specifically limit this.

[0036] 102. Based on the initial search information, determine the candidate display objects among multiple objects to be displayed.

[0037] In implementation, the platform's database stores multiple objects to be displayed. After obtaining initial search information, a search can be performed on these objects to identify one or more candidate objects that match the initial search information.

[0038] In this embodiment of the disclosure, the method for search retrieval based on initial search information can be any reasonable method. For example, the initial search information and the display information of multiple objects to be displayed stored in the database can be input into the matching model to obtain the matching degree between the initial search information and each object to be displayed. Multiple objects to be displayed with a matching degree greater than the matching degree threshold can be determined as candidate display objects. Alternatively, the object to be displayed corresponding to the largest preset number of matching degrees among multiple matching degrees can be determined as a candidate display object, and so on.

[0039] 103. Based on the display information of multiple information units and multiple candidate display objects, determine the importance score of each information unit to the display information of each candidate display object.

[0040] The display information of the candidate display object is a detailed description of the candidate display object. For example, when the candidate display object is a product, its corresponding display information may include the product's title information, detailed description information, image information, etc.

[0041] In implementation, after identifying candidate display objects, each information unit can be processed as follows: calculate the contribution score of each information unit to recalling the candidate display object, i.e., the importance score, which reflects the importance of the information unit in the initial search information.

[0042] In this embodiment of the disclosure, the method for determining the importance score can be any reasonable method. For example, an attribution value algorithm can be used to calculate the attribution value of each information unit for the display information of each candidate display object as the importance score. The attribution value can also include positive attribution value and / or negative attribution value.

[0043] For example, Integrated Gradients (IG) can be used to calculate the display information of this information unit and multiple candidate display objects, determining the positive and negative attribution values ​​of this information unit for the display information of each candidate display object. These positive and negative attribution values ​​reflect the degree of contribution of this information unit to recalling the candidate display object. The larger the absolute value of the positive or negative attribution value corresponding to this information unit, the more critical the information unit is.

[0044] When the positive attribution value corresponding to the information unit and the candidate display object is greater than the first positive attribution value threshold, and the negative attribution value corresponding to the information unit is greater than the first negative attribution value threshold (i.e., the absolute value of the negative attribution value is less than the absolute value of the first negative attribution value threshold), it indicates that the information unit makes a significant contribution to recalling the candidate display object.

[0045] Similarly, when the positive attribution value corresponding to the information unit and the candidate display object is less than the second positive attribution value threshold, and the negative attribution value corresponding to the information unit is less than the second negative attribution value threshold (i.e., the absolute value of the negative attribution value is greater than the absolute value of the second negative attribution value threshold), it indicates that the information unit also makes a significant contribution to recalling the candidate display object.

[0046] Among them, the first positive attribution value threshold is greater than the second positive attribution value threshold, and the first negative attribution value threshold is greater than the second recurrence attribution value threshold (that is, the absolute value of the first negative attribution value threshold is less than the absolute value of the second recurrence attribution value threshold).

[0047] For example, the initial search information is "running shoes not leather waterproof". The positive attribution value A+ and the negative attribution value A- for "waterproof", "not", and "leather" for the same candidate display object are calculated as follows: "waterproof": A+=0.85,A =-0.02 indicates that the information unit "waterproof" has a strong positive attribute for this candidate display object.

[0048] “not”: A+=0.01, A = -0.78, which means that the information unit "not" has a strong negative signal for the candidate display object (it is a key exclusion word).

[0049] “leather”: A+=0.12, A =-0.65, this information unit is an excluded object modified by "not", and has a strong negative signal.

[0050] In this way, while improving the information units for strong positive attributes, it also improves the information units for strong negative attributes. This avoids the situation in related technologies where the information unit might be ignored due to the low positive attribution of "leather," leading to the recall of leather shoes. On the contrary, the method provided in this embodiment can retain "not leather" as a complete constraint unit, thereby improving the accuracy of rewriting.

[0051] 104. Based on the initial search information, the importance score of each information unit, and the rewriting model, determine the rewriting information corresponding to the initial search information.

[0052] In practice, the trained rewriting model can rewrite the initial search information based on the initial search information and the importance score value corresponding to each information unit, thereby obtaining the rewritten information corresponding to the initial search information.

[0053] For example, if the initial search information is "that popular sneaker", the rewritten information could be "this year's new men's breathable running shoes".

[0054] In the above rewriting process, the importance score can guide the rewriting model to identify which information units have a greater impact on the accurate recall of display objects and are more critical. This helps the rewriting model focus on the key information to obtain more accurate rewriting information, thereby improving the accuracy of the identified display objects.

[0055] 105. Based on the rewritten information, determine the display object from among multiple objects to be displayed, and send the display information of the display object to the target account.

[0056] In practice, once the rewritten information is determined, the display object to be shown to the user can be re-determined from multiple objects to be displayed based on the rewritten information. This display object is obtained based on the rewritten information that better expresses the user's intent. Therefore, this display object is more in line with the user's search intent than the candidate display objects determined based on the initial search information. Sending the display information of this display object to the target account for display can improve the user's search experience.

[0057] In this embodiment of the disclosure, there are multiple methods for determining candidate display objects based on initial search information in step 102 above. Several possible methods are described below: In one possible implementation, the method for determining candidate display objects based on initial search information can be: based on initial search information, multiple objects to be displayed, and a semantic matching model, determine candidate display objects from multiple objects to be displayed.

[0058] In implementation, features can be extracted from the initial search information and the display information of each object to be displayed, respectively, to obtain the search feature vector corresponding to the initial search information and the display feature vector corresponding to each object to be displayed.

[0059] Then, the search feature vector and the display feature vector corresponding to each object to be displayed are input into the semantic matching model to obtain the semantic matching degree between the search feature vector and the display feature vector corresponding to each object to be displayed. Then, the top four number of objects to be displayed with the highest semantic matching degree are selected as candidate display objects.

[0060] The fourth number can be any reasonable value, such as 20, 15, etc.

[0061] In this embodiment of the disclosure, the semantic matching model can be any reasonable computational model or machine learning model, for example, it can be a BGE (BAAI General Embedding-M3, Beijing Academy of Artificial Intelligence General Embedding Model)-M3 (Multi-Functionality + Multi-Linguality + Multi-Granularity, Multi-functionality + Multi-linguality + Multi-granularity) model.

[0062] In one possible implementation, the method for determining candidate display objects based on initial search information can also be: based on initial search information, multiple objects to be displayed, and a keyword matching model, determine candidate display objects from multiple objects to be displayed.

[0063] In implementation, the keyword information from the initial search information and the display information of multiple objects to be displayed can be input into the keyword matching model to obtain the keyword matching degree between the output initial search information and each object to be displayed. Then, the top five objects with the highest matching degree are selected as candidate display objects.

[0064] The fifth number can be any reasonable value, such as 20, 15, etc.

[0065] In this embodiment of the disclosure, the keyword matching model can be any reasonable computational model or machine learning model, for example, it can be the BM25 (Best Matching 25) model.

[0066] In one possible implementation, see Figure 3 Another method for determining candidate display objects based on initial search information is as follows: Based on the initial search information, multiple objects to be displayed, and a semantic matching model, determine the first candidate display object corresponding to the initial search information among the multiple objects to be displayed. Based on the initial search information, multiple objects to be displayed, and a keyword matching model, determine the second candidate display object corresponding to the initial search information among the multiple objects to be displayed. Based on the first and second candidate display objects, determine the final candidate display object.

[0067] In implementation, two-way recall can be performed in parallel based on the initial search information, namely, two-way recall through semantic matching and keyword matching to obtain the first candidate display object and the second candidate display object respectively. Then, one or more of the first candidate display object and the second candidate display object can be selected as candidate display objects.

[0068] In this way, the resulting candidate display objects comprehensively consider both semantic recall and keyword recall, improving the comprehensive coverage of the positive and negative attribution values ​​of the subsequently determined candidate display objects, thereby improving the accuracy of the determined positive and negative attribution values.

[0069] In one possible implementation, the method for determining the first candidate display object based on the semantic matching model described above can be as follows: based on initial search information, multiple objects to be displayed, and the semantic matching model, determine the semantic matching degree between the initial search information and each object to be displayed. Then, based on the semantic matching degree, determine the first candidate display object corresponding to the initial search information among the multiple objects to be displayed.

[0070] In implementation, the initial search information and the display information of multiple objects to be displayed are input into the semantic matching model to obtain the semantic matching degree between the output initial search information and each object to be displayed. Then, multiple objects to be displayed with a semantic matching degree greater than the semantic matching degree threshold can be determined as the first candidate display objects. Alternatively, the object to be displayed corresponding to the largest first preset number of semantic matching degrees among multiple semantic matching degrees can be determined as the first candidate display object.

[0071] The first preset number can be any reasonable value, such as 10, 20, etc.

[0072] In one possible implementation, the method for determining the second candidate display object based on the keyword matching model described above can be as follows: based on the initial search information, multiple objects to be displayed, and the keyword matching model, determine the keyword matching degree between the initial search information and each object to be displayed. Then, based on the keyword matching degree, determine the second candidate display object corresponding to the initial search information among the multiple objects to be displayed.

[0073] In implementation, the keyword information of the initial search information and the display information of multiple objects to be displayed are input into the keyword matching model to obtain the keyword matching degree between the output initial search information and each object to be displayed. Then, multiple objects to be displayed with a keyword matching degree greater than the keyword matching degree threshold can be determined as the second candidate display objects. Alternatively, the object to be displayed corresponding to the largest preset number of keyword matching degrees among multiple keyword matching degrees can be determined as the second candidate display object.

[0074] The second preset number can be any reasonable value, such as 5, 30, etc.

[0075] After determining the first candidate display object and the second candidate display object, the following processing can be performed: based on the semantic matching degree corresponding to the first candidate display object and the keyword matching degree corresponding to the second candidate display object, a candidate display object is determined from the first candidate display object and the second candidate display object.

[0076] In practice, based on the semantic matching degree and keyword matching degree, the candidate display object with a higher semantic matching degree and keyword matching degree can be determined from the first candidate display object and the second candidate display object. This improves the accuracy of the determined candidate display object and the accuracy of the determined importance score.

[0077] In one possible implementation, the method for determining candidate display objects based on the semantic matching degree of the first candidate display object and the keyword matching degree of the second candidate display object can be as follows: First, determine first ranking information of the first candidate display object based on the semantic matching degree. Second, determine second ranking information of the second candidate display object based on the keyword matching degree. Third, determine third ranking information for each object to be displayed in the display object set composed of the first and second candidate display objects based on the first and second ranking information. Finally, determine candidate display objects from the display object set based on the third ranking information.

[0078] In implementation, the first candidate display objects can be sorted in descending order of semantic matching degree to obtain the first ranking information. The second candidate display objects can be sorted in descending order of keyword matching degree to obtain the second ranking information. Then, based on the first and second ranking information, the first and second candidate display objects are comprehensively ranked, or merged, to obtain the third ranking information. This third ranking information reflects the comprehensive matching degree between each object to be displayed in the set of display objects and the initial search information. The candidate display objects determined based on this third ranking information have a high semantic matching degree and keyword matching degree with the initial search information.

[0079] In one possible implementation, the method for determining the third ranking information based on the first and second ranking information described above could be as follows: Based on the first ranking information, the second ranking information, and RRF (Reciprocal Rank Fusion), determine the fusion score corresponding to each object to be displayed in the set of displayed objects. Based on the fusion score, determine the third ranking information.

[0080] In implementation, the first and second sorting information can be input into the RRF algorithm to obtain the fusion score corresponding to each object to be displayed in the display object set. The objects to be displayed in the display object set can be sorted in descending order of fusion score to obtain the third sorting information.

[0081] In one possible implementation, the method for determining candidate display objects in the display object set based on the third ranking information described above can be as follows: If the fusion score of the first display object in the third ranking information is greater than a first score threshold, then the first number of display objects in the third ranking information are determined as candidate display objects, where the third ranking information is arranged in descending order of fusion score. If the fusion score of the first display object in the third ranking information is less than a second score threshold, then the second number of display objects in the third ranking information are determined as candidate display objects, where the second score threshold is less than the first score threshold. If the fusion score of the first display object in the third ranking information is not less than the second score threshold and not greater than the first score threshold, then the third number of display objects in the third ranking information are determined as candidate display objects, where the third number is greater than the first number and less than the second number.

[0082] In practice, the maximum value in the fusion score can be compared with the first score threshold and the second score threshold. If the fusion score is greater than the first score threshold, it means that the recalled objects to be displayed are more accurate, and the number of candidate objects to be displayed can be reduced. Only the first few objects to be displayed are needed to reflect the attribution situation.

[0083] If the fusion score is less than the second score threshold, it indicates that the recalled objects to be displayed are not very accurate. In this case, the number of selected candidate objects can be increased. This method of increasing the number can improve the accuracy and comprehensiveness of the subsequently determined positive and negative attribution values.

[0084] If the fusion score is not less than the second score threshold and not greater than the first score threshold, it indicates that the accuracy of the recalled objects to be displayed is moderate. In this case, a number of candidate objects that is greater than the first number and less than the second number can be selected to determine a more accurate and comprehensive positive and negative attribution value.

[0085] In this embodiment of the disclosure, the first number, the second number, and the third number can be any reasonable value. For example, the first number can be 3, the second number can be 20, and the third number can be 10.

[0086] In this embodiment of the disclosure, there are multiple methods for determining the rewrite information corresponding to the initial search information in step 104 above. Several possible methods are described below: In one possible implementation, see Figure 4The method for determining rewritten information can be as follows: For each information unit, determine the positive attribution score of the information unit based on the positive attribution value of the information unit's display information for each candidate display object, and determine the negative attribution score of the information unit based on the negative attribution value of the information unit's display information for each candidate display object. The rewritten information is determined based on the initial search information, the positive and negative attribution scores of each information unit, and the rewriting model.

[0087] In practice, since multiple candidate display objects are identified, for each information unit, an information unit has a positive attribution value for multiple candidate display objects. Therefore, the positive attribution score of the information unit can be calculated based on the positive attribution values ​​for these multiple candidate display objects. This positive attribution score can evaluate the contribution of the information unit to the recall from a more comprehensive perspective.

[0088] Similarly, for each information unit, an information unit has a negative attribution value for multiple candidate display objects. Therefore, the negative attribution score of the information unit can be calculated based on the negative attribution values ​​of these multiple candidate display objects. This negative attribution score can also evaluate the contribution of the information unit to the recall from a more comprehensive perspective.

[0089] Then, the initial search information, the positive attribution score and the negative attribution score of each information unit can be input into the rewriting model to guide the rewriting model to identify that information units with larger positive and negative attribution scores have a greater impact on the accurate recall of display objects and are more critical. This helps the rewriting model focus on the key information to output more accurate rewriting information, thereby improving the accuracy of the identified display objects.

[0090] In one possible implementation, the method for determining the positive attribution score in the above method can be: based on the fusion score corresponding to each candidate display object, perform a weighted average of the positive attribution values ​​of the information unit for the display information of each candidate display object to obtain the positive attribution score of the information unit.

[0091] In practice, candidate display objects with higher fusion scores are more in line with the recall criteria of the initial search information. Therefore, the positive attribution value of each candidate display object can be weighted by the fusion score, so that the positive attribution value of the candidate display object that is more in line with the recall criteria of the initial search information has a higher weight. This makes the positive attribution score obtained after weighted averaging more accurately reflect the importance of the information unit.

[0092] Similarly, a method for determining negative attribution scores could be: based on the fusion score corresponding to each candidate display object, perform a weighted average of the absolute values ​​of the negative attribution values ​​of the information unit for the display information of each candidate display object, and obtain the negative attribution score of the information unit.

[0093] In practice, candidate display objects with higher fusion scores are more in line with the recall criteria of the initial search information. Therefore, the negative attribution value of each candidate display object can be weighted by the fusion score, so that the negative attribution value of the candidate display object that is more in line with the recall criteria of the initial search information has a higher weight. This makes the negative attribution score obtained after weighted averaging more accurately reflect the importance of the information unit.

[0094] In one possible implementation, the method for sending display information provided in this disclosure embodiment may further include the following processing: determining the target attribute information corresponding to each information unit from the multiple attribute information based on multiple information units and multiple attribute information corresponding to multiple objects to be displayed.

[0095] In practice, the platform's database pre-stores multiple attribute information, such as the product category, color, size, price range, material, etc. This disclosure does not limit this aspect.

[0096] One or more attribute information corresponding to each information unit can be determined, which is the target attribute information.

[0097] In the embodiments of this disclosure, the method for determining the target attribute information can be any reasonable method. For example, the target attribute information corresponding to each information unit can be determined based on multiple information units, multiple attribute information corresponding to multiple objects to be displayed, and an attribute matching model. The attribute matching model can be any reasonable model structure, such as a NER (Named Entity Recognition) model, etc.

[0098] Correspondingly, the method for determining the rewriting information is as follows: based on the initial search information, the importance score value corresponding to each information unit, the target attribute information corresponding to each information unit, and the rewriting model, the rewriting information corresponding to the initial search information is determined.

[0099] In implementation, the initial search information, the importance score of each information unit, and the target attribute information of each information unit can be input into the rewriting model to guide the importance level and target attributes of different information units in the rewriting model, so as to make importance judgments and semantic rewriting, and obtain more accurate rewriting information output.

[0100] In one possible implementation, the method for sending display information provided in this disclosure embodiment may further include the following processing: for each target attribute information, determining the attribute importance score corresponding to each target attribute information based on the importance score value of the information unit corresponding to the target attribute information.

[0101] In practice, for each target attribute information, the statistical value of the importance score of one or more information units corresponding to that target attribute information can be calculated, that is, the attribute importance score corresponding to that target attribute information can be obtained. The attribute importance score can reflect the degree of contribution of the target attribute information to the recall in the initial search information.

[0102] Correspondingly, the method for determining the rewriting information is as follows: based on the initial search information, the importance score value corresponding to each information unit, the target attribute information corresponding to each information unit, the attribute importance score value corresponding to each target attribute information, and the rewriting model, the rewriting information corresponding to the initial search information is determined.

[0103] In implementation, the initial search information, the importance score of each information unit, the target attribute information of each information unit, and the attribute importance score of each target attribute information can all be input into the rewriting model to obtain more accurate rewriting information that is more effective for recall.

[0104] In one possible implementation, see Figure 5 The method for determining the attribute importance score corresponding to each target attribute information can be as follows: For each target attribute information, based on the positive attribution value of the information unit corresponding to the target attribute information, determine the positive attribution attribute value corresponding to the target attribute information; based on the negative attribution value of the information unit corresponding to the target attribute information, determine the negative attribution attribute value corresponding to the target attribute information.

[0105] In implementation, for each target attribute information, the statistical value of one or more positive attribution values ​​corresponding to that target attribute information can be calculated, thus obtaining the positive attribution attribute value for that target attribute information. Similarly, the statistical value of one or more negative attribution values ​​corresponding to that target attribute information can be calculated, thus obtaining the negative attribution attribute value for that target attribute information. These positive and negative attribution attribute values ​​can reflect the degree of contribution of the target attribute information, as presented in the initial search information, to the recall.

[0106] Correspondingly, see Figure 5The method for determining the rewriting information is as follows: based on the initial search information, the positive and negative attribution values ​​corresponding to each information unit, the target attribute information corresponding to each information unit, the positive and negative attribution attribute values ​​corresponding to each target attribute information, and the rewriting model, the rewriting information corresponding to the initial search information is determined.

[0107] In implementation, the initial search information, the positive and negative attribution values ​​corresponding to each information unit, the target attribute information corresponding to each information unit, and the positive and negative attribution attribute values ​​corresponding to each target attribute information can all be input into the rewriting model to obtain more accurate rewriting information that is more effective for recall.

[0108] In one possible implementation, the method for determining the target attribute information corresponding to each information unit in the above method may be: based on the initial search information, multiple information units, multiple attribute information corresponding to multiple objects to be displayed, and the attribute matching model, determine the attribute matching degree between each information unit and each object to be displayed, and determine the target attribute information corresponding to each information unit based on the attribute matching degree between each information unit and each object to be displayed.

[0109] In implementation, the initial search information, multiple information units, and multiple attribute information corresponding to multiple objects to be displayed are input into the attribute matching model to obtain the attribute matching degree between each information unit and each object to be displayed.

[0110] Then, for each information unit, the following processing can be performed: among the multiple attribute matching degrees corresponding to the information unit, the attribute information corresponding to one or more attribute matching degrees that are greater than the attribute matching degree threshold is determined as the target attribute information corresponding to the information unit.

[0111] Alternatively, among the multiple attribute matching degrees corresponding to the information unit, the attribute information corresponding to the third preset number of attribute matching degrees with the highest attribute matching degree is determined as the target attribute information corresponding to the information unit. The third preset number can be any reasonable value, and this embodiment does not limit it.

[0112] Correspondingly, the method for determining the positive attribution attribute value corresponding to the target attribute information in the above method can be: based on the attribute matching degree corresponding to the target attribute information, perform weighted average processing on the positive attribution value of the information unit corresponding to the target attribute information to obtain the positive attribution attribute value corresponding to the target attribute information.

[0113] The method described above for determining the negative attribution attribute value corresponding to the target attribute information can be as follows: based on the attribute matching degree corresponding to the target attribute information, perform a weighted average of the absolute values ​​of the negative attribution values ​​of the information units corresponding to the target attribute information to obtain the negative attribution attribute value corresponding to the target attribute information.

[0114] In this disclosure embodiment, see Figure 6 After determining the rewrite information, the following processing can be performed: The rewrite information and / or the display object are validated based on preset validation rules. If the rewrite information does not conform to the preset validation rules, the rewrite information is updated, and the display object is determined based on the updated rewrite information.

[0115] In this way, by verifying the rewritten information and / or the displayed object, it can be determined whether the rewritten information is accurate. If it does not meet the preset verification rules, it means that the rewritten information does not achieve the expected accuracy and can be updated and adjusted.

[0116] In one possible implementation, the verification process may include: validating the format of the rewritten information.

[0117] Correspondingly, if the rewritten information does not conform to the format validity, the rewritten information can be updated based on the initial search information, the importance score value corresponding to each information unit, and the rewriting model.

[0118] That is, if the rewritten information does not conform to the format validity, it means that a data anomaly has occurred during the processing. At this time, the initial search information can be rewritten again based on the initial search information, the importance score value corresponding to each information unit, and the rewriting model to obtain new rewritten information.

[0119] Then, the new rewritten information can be validated again. If the validation still does not meet the preset validation rules, the candidate display object can be directly determined as the final display object, and the display information of the candidate display object can be sent to the target account.

[0120] In one possible implementation, the verification process may further include: verifying whether the preset reserved attributes present in the initial search information exist in the rewritten information based on preset reserved attributes among multiple attribute information.

[0121] Specifically, one or more attributes can be selected as preset retained attributes from multiple attribute information sources, based on requirements. That is, when the rewriting model rewrites the initial search information, it can be told that the target attribute information is a preset retained attribute and must be preserved, not deleted. This ensures that semantically clear information pointing to specific needs is preserved from the initial search information, avoiding semantic loss due to the deletion of such information. For example, color attributes (corresponding information units could be red, yellow, etc.) can be used as preset retained attributes.

[0122] Correspondingly, if the preset retained attributes present in the initial search information do not exist in the rewritten information, it means that the rewritten information does not have the information corresponding to the preset retained attributes. In this case, the rewritten information can be updated based on the initial search information, the positive attribution value and the corresponding negative attribution value of each information unit, and the rewriting model.

[0123] That is, if the rewritten information does not have the information corresponding to the preset retention attribute, it means that a data anomaly has occurred during the processing. At this time, the initial search information can be rewritten again based on the initial search information, the importance score value corresponding to each information unit, and the rewriting model to obtain new rewritten information.

[0124] Then, the new rewritten information can be validated again. If the validation still does not meet the preset validation rules, the candidate display object can be directly determined as the final display object, and the display information of the candidate display object can be sent to the target account.

[0125] In one possible implementation, the verification process may further include determining whether at least one display object exists. That is, if at least one display object cannot be determined from the objects to be displayed based on the rewritten information, then recall display cannot be provided to the user. In this case, based on the preset extended attributes among multiple attribute information, the information corresponding to the preset extended attributes existing in the rewritten information can be deleted to obtain the updated rewritten information.

[0126] In practice, the rewriting model will also output the attribute information corresponding to the multiple rewriting information units included in the rewriting information. If at least one display object cannot be determined from the objects to be displayed based on the rewriting information, the information corresponding to the preset extended attributes existing in the rewriting information can be deleted.

[0127] The preset extended attribute is the attribute information that the rewriting model is instructed to perform extended rewriting in the prompt words input into the rewriting model during the rewriting process. For example, the price range can be a preset extended attribute. When the rewriting model rewrites the initial search information, if there is an information unit with the price range attribute in the initial search information, the rewriting model will perform extended rewriting on the information unit with the price range attribute.

[0128] Therefore, if at least one display object cannot be determined from the objects to be displayed based on the rewritten information, the information corresponding to these preset extended attributes can be deleted in turn to expand the recall scope and improve the recall effect.

[0129] In one possible implementation, if the displayed object still cannot be determined based on the updated rewritten information, then based on the preset transformation attributes among multiple attribute information, the information corresponding to the preset transformation attributes in the previously updated rewritten information is restored to its original information, resulting in the updated rewritten information again. Here, the original information corresponding to the information corresponding to the preset transformation attribute is the information unit corresponding to that preset transformation attribute in the initial search information. Based on the updated rewritten information, the displayed object is determined from among the multiple objects to be displayed.

[0130] In practice, the preset conversion attribute is the attribute information in the prompt words input to the rewriting model during the rewriting process that instructs the rewriting model to convert and rewrite similar words. That is, if the information unit is a preset conversion attribute, the rewriting model can determine the conversion information corresponding to the information unit based on the conversion correspondence table input to the rewriting model, and write the conversion information into the rewriting information for replacement, so as to make the semantics clearer.

[0131] However, if at least one display object cannot be determined from the objects to be displayed based on the rewritten information after the last update, the information corresponding to the preset transformation attributes in the rewritten information can be restored to the original information.

[0132] Try using the original information for recall to improve recall effectiveness.

[0133] In one possible implementation, if the target audience still cannot be determined based on the updated rewrite information described above, then display information for at least one candidate target audience is sent to the target account. That is, if the target audience still cannot be determined using the above method, one or more candidate target audiences can be directly selected as the target audience and sent to the target account for display.

[0134] In this way, by using one or more of the above verification methods and corresponding update methods, the accuracy of the determined rewrite information is improved, which in turn improves the accuracy of the determined display objects, thereby enhancing the user's search experience.

[0135] All of the above-mentioned optional technical solutions can be combined in any way to form optional embodiments of this disclosure, and will not be described in detail here.

[0136] The scheme mentioned in this embodiment can identify candidate display objects that match the initial search information. These candidate display objects initially reflect the intent expressed by the initial search information. Then, the importance score of each information unit of the initial search information to each candidate display object can be calculated. This importance score reflects the contribution of the word unit to recalling the candidate display object. Then, based on the initial search information, the importance score of each information unit to each candidate display object, and the rewriting model, the rewritten information corresponding to the initial search information can be obtained. In the above rewriting process, the importance score can guide the rewriting model to identify which information units have a greater impact on accurately recalling display objects, thereby helping the rewriting model focus on the key information to obtain more accurate rewritten information, thereby improving the accuracy of the identified display objects.

[0137] This disclosure provides an apparatus for sending display information, which may be the computer device described in the above embodiments, such as... Figure 7 As shown, the device includes: The acquisition module 710 is used to acquire the initial search information input by the target account and the multiple information units included in the initial search information; The first determining module 720 is used to determine candidate display objects among the plurality of objects to be displayed based on the initial search information; The second determining module 730 is used to determine the importance score of each information unit to the display information of each candidate display object based on the display information of the plurality of information units and the plurality of candidate display objects. The third determining module 740 is used to determine the rewriting information corresponding to the initial search information based on the initial search information, the importance score value corresponding to each information unit, and the rewriting model. The sending module 750 is used to determine the display object among the plurality of objects to be displayed based on the rewritten information, and send the display information of the display object to the target account.

[0138] In one possible implementation, the importance score includes positive attribution values ​​and / or negative attribution values.

[0139] In one possible implementation, the first determining module 720 is configured to: Based on the initial search information, the plurality of objects to be displayed, and the semantic matching model, a first candidate display object corresponding to the initial search information is determined from the plurality of objects to be displayed. Based on the initial search information, the plurality of objects to be displayed, and the keyword matching model, a second candidate object corresponding to the initial search information is determined from the plurality of objects to be displayed. The candidate display object is determined based on the first candidate display object and the second candidate display object.

[0140] In one possible implementation, the first determining module 720 is configured to: Based on the initial search information, the multiple objects to be displayed, and the semantic matching model, determine the semantic matching degree between the initial search information and each object to be displayed; Based on the semantic matching degree, the first candidate display object corresponding to the initial search information is determined among the multiple objects to be displayed; Based on the initial search information, the multiple objects to be displayed, and the keyword matching model, determine the keyword matching degree between the initial search information and each object to be displayed; Based on the keyword matching degree, a second candidate display object corresponding to the initial search information is determined from the plurality of objects to be displayed; Based on the semantic matching degree corresponding to the first candidate display object and the keyword matching degree corresponding to the second candidate display object, the candidate display object is determined from the first candidate display object and the second candidate display object.

[0141] In one possible implementation, the first determining module 720 is configured to: The first ranking information of the first candidate display object is determined based on the semantic matching degree; The second ranking information of the second candidate display object is determined based on the keyword matching degree; Based on the first sorting information and the second sorting information, the third sorting information corresponding to each object to be displayed in the set of display objects composed of the first candidate display objects and the second candidate display objects is determined; Based on the third sorting information, the candidate display objects are determined in the set of display objects.

[0142] In one possible implementation, the second determining module 730 is further configured to: Based on the multiple information units and the multiple attribute information corresponding to the multiple objects to be displayed, the target attribute information corresponding to each information unit is determined from the multiple attribute information; The third determining module 740 is used for: Based on the initial search information, the importance score corresponding to each information unit, the target attribute information corresponding to each information unit, and the rewriting model, the rewriting information corresponding to the initial search information is determined.

[0143] In one possible implementation, the second determining module 730 is further configured to: For each target attribute information, based on the importance score of the information unit corresponding to the target attribute information, the attribute importance score corresponding to each target attribute information is determined; The third determining module 740 is used for: Based on the initial search information, the importance score corresponding to each information unit, the target attribute information corresponding to each information unit, the attribute importance score corresponding to each target attribute information, and the rewriting model, the rewriting information corresponding to the initial search information is determined.

[0144] In one possible implementation, the sending module 750 is further configured to: The rewritten information and / or the displayed object are verified based on preset verification rules. If the rewritten information does not conform to the preset verification rules, the rewritten information is updated, and the display object is determined based on the updated rewritten information.

[0145] It should be noted that the device for sending display information provided in the above embodiments is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the device for sending display information and the method embodiment for sending display information provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiment, which will not be repeated here.

[0146] Figure 8 A structural block diagram of a terminal 800 provided in an exemplary embodiment of this disclosure is shown. The terminal can be a computer device as described in the above embodiments. The terminal 800 can be a smartphone, tablet computer, MP3 player (moving picture experts group audio layer III), MP4 player (moving picture experts group audio layer IV), laptop computer, or desktop computer. The terminal 800 may also be referred to as a user device, portable terminal, laptop terminal, desktop terminal, or other names.

[0147] Typically, terminal 800 includes a processor 801 and a memory 802.

[0148] Processor 801 may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. Processor 801 may be implemented using at least one hardware form selected from DSP (digital signal processing), FPGA (field-programmable gate array), and PLA (programmable logic array). Processor 801 may also include a main processor and a coprocessor. The main processor, also known as a CPU (central processing unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 801 may integrate a GPU (graphics processing unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 801 may also include an AI (artificial intelligence) processor, which is used to handle computational operations related to machine learning.

[0149] The memory 802 may include one or more computer-readable storage media, which may be non-transitory. The memory 802 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 802 are used to store at least one instruction, which is executed by the processor 801 to implement the method of sending display information provided in the method embodiments of this disclosure.

[0150] In some embodiments, the terminal 800 may also optionally include a peripheral device interface 803 and at least one peripheral device. The processor 801, memory 802, and peripheral device interface 803 can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface 803 via a bus, signal line, or circuit board. Specifically, the peripheral device includes at least one of the following: a radio frequency circuit 804, a display screen 805, a camera 806, an audio circuit 807, a positioning component 808, and a power supply 809.

[0151] Peripheral device interface 803 can be used to connect at least one I / O (input / output) related peripheral device to processor 801 and memory 802. In some embodiments, processor 801, memory 802 and peripheral device interface 803 are integrated on the same chip or circuit board; in some other embodiments, any one or two of processor 801, memory 802 and peripheral device interface 803 can be implemented on separate chips or circuit boards, which is not limited in this embodiment.

[0152] The radio frequency (RF) circuit 804 is used to receive and transmit RF (radio frequency) signals, also known as electromagnetic signals. The RF circuit 804 communicates with communication networks and other communication devices via electromagnetic signals. The RF circuit 804 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals back into electrical signals. Optionally, the RF circuit 804 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, etc. The RF circuit 804 can communicate with other terminals through at least one wireless communication protocol. This wireless communication protocol includes, but is not limited to: metropolitan area networks (MANs), various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area networks (WLANs), and / or WiFi (wireless fidelity) networks. In some embodiments, the RF circuit 804 may also include circuitry related to NFC (near field communication), which is not limited in this disclosure.

[0153] Display screen 805 is used to display a user interface (UI). This UI may include graphics, text, icons, videos, and any combination thereof. When display screen 805 is a touch display screen, it also has the ability to collect touch signals on or above its surface. These touch signals can be input as control signals to processor 801 for processing. In this case, display screen 805 can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments, there may be one display screen 805, which serves as the front panel of terminal 800; in other embodiments, there may be at least two display screens, respectively disposed on different surfaces of terminal 800 or in a folded design; in still other embodiments, display screen 805 may be a flexible display screen, disposed on a curved or folded surface of terminal 800. Furthermore, display screen 805 may be configured as a non-rectangular, irregular shape, i.e., a non-rectangular screen. Display screen 805 may be made of materials such as LCD (liquid crystal display) or OLED (organic light-emitting diode).

[0154] The camera assembly 806 is used to acquire images or videos. Optionally, the camera assembly 806 includes a front-facing camera and a rear-facing camera. Typically, the front-facing camera is located on the front panel of the terminal, and the rear-facing camera is located on the back of the terminal. In some embodiments, there are at least two rear-facing cameras, which are any one of a main camera, a depth-sensing camera, a wide-angle camera, and a telephoto camera, to achieve background blurring by fusion of the main camera and the depth-sensing camera, panoramic shooting by fusion of the main camera and the wide-angle camera, VR (virtual reality) shooting, or other fusion shooting functions. In some embodiments, the camera assembly 806 may also include a flash. The flash can be a single-color temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm-light flash and a cool-light flash, which can be used for light compensation at different color temperatures.

[0155] The audio circuit 807 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, converting the sound waves into electrical signals that are input to the processor 801 for processing, or input to the radio frequency circuit 804 to achieve voice communication. For stereo sound acquisition or noise reduction purposes, multiple microphones may be used, each located at a different part of the terminal 800. The microphone may also be an array microphone or an omnidirectional microphone. The speaker is used to convert the electrical signals from the processor 801 or the radio frequency circuit 804 into sound waves. The speaker may be a conventional diaphragm speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can convert electrical signals not only into audible sound waves but also into inaudible sound waves for purposes such as distance measurement. In some embodiments, the audio circuit 807 may also include a headphone jack.

[0156] The positioning component 808 is used to determine the current geographic location of the positioning terminal 800 in order to enable navigation or LBS (location-based service). The positioning component 808 can be a positioning component based on GPS (global positioning system), BeiDou system, Grenadin system, or Galileo system.

[0157] Power supply 809 is used to supply power to the various components in terminal 800. Power supply 809 can be AC ​​power, DC power, a disposable battery, or a rechargeable battery. When power supply 809 includes a rechargeable battery, the rechargeable battery can support wired or wireless charging. The rechargeable battery can also be used to support fast charging technology.

[0158] In some embodiments, the terminal 800 further includes one or more sensors 810. The one or more sensors 810 include, but are not limited to: an accelerometer 811, a gyroscope 812, a pressure sensor 813, a fingerprint sensor 814, an optical sensor 815, and a proximity sensor 816.

[0159] Accelerometer 811 can detect the magnitude of acceleration on the three coordinate axes of a coordinate system established by terminal 800. For example, accelerometer 811 can be used to detect the components of gravitational acceleration on the three coordinate axes. Processor 801 can control display screen 805 to display the user interface in either a landscape or portrait view based on the gravitational acceleration signal acquired by accelerometer 811. Accelerometer 811 can also be used for games or for acquiring user motion data.

[0160] The gyroscope sensor 812 can detect the orientation and rotation angle of the terminal 800. The gyroscope sensor 812, in conjunction with the accelerometer sensor 811, can collect 3D motion data from the user on the terminal 800. Based on the data collected by the gyroscope sensor 812, the processor 801 can perform the following functions: motion sensing (e.g., changing the UI based on the user's tilt), image stabilization during shooting, game control, and inertial navigation.

[0161] The pressure sensor 813 can be disposed on the side bezel of the terminal 800 and / or on the lower layer of the display screen 805. When the pressure sensor 813 is disposed on the side bezel of the terminal 800, it can detect the user's grip signal on the terminal 800, and the processor 801 can perform left / right hand recognition or quick operation based on the grip signal collected by the pressure sensor 813. When the pressure sensor 813 is disposed on the lower layer of the display screen 805, the processor 801 can control the operable controls on the UI interface based on the user's pressure operation on the display screen 805. The operable controls include at least one of button controls, scroll bar controls, icon controls, and menu controls.

[0162] The fingerprint sensor 814 is used to collect the user's fingerprint. The processor 801 identifies the user's identity based on the fingerprint collected by the fingerprint sensor 814, or the fingerprint sensor 814 identifies the user's identity based on the collected fingerprint. When the user's identity is identified as trusted, the processor 801 authorizes the user to perform relevant sensitive operations, including unlocking the screen, viewing encrypted information, downloading software, making payments, and changing settings. The fingerprint sensor 814 can be located on the front, back, or side of the terminal 800. When the terminal 800 has physical buttons or a manufacturer's logo, the fingerprint sensor 814 can be integrated with the physical buttons or manufacturer's logo.

[0163] An optical sensor 815 is used to collect ambient light intensity. In one embodiment, the processor 801 can control the display brightness of the display screen 805 based on the ambient light intensity collected by the optical sensor 815. Specifically, when the ambient light intensity is high, the display brightness of the display screen 805 is increased; when the ambient light intensity is low, the display brightness of the display screen 805 is decreased. In another embodiment, the processor 801 can also dynamically adjust the shooting parameters of the camera assembly 806 based on the ambient light intensity collected by the optical sensor 815.

[0164] The proximity sensor 816, also known as a distance sensor, is typically located on the front panel of the terminal 800. The proximity sensor 816 is used to detect the distance between the user and the front of the terminal 800. In one embodiment, when the proximity sensor 816 detects that the distance between the user and the front of the terminal 800 is gradually decreasing, the processor 801 controls the display screen 805 to switch from a screen-on state to a screen-off state; when the proximity sensor 816 detects that the distance between the user and the front of the terminal 800 is gradually increasing, the processor 801 controls the display screen 805 to switch from a screen-off state to a screen-on state.

[0165] Those skilled in the art will understand that Figure 8 The structure shown does not constitute a limitation on terminal 800 and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0166] Figure 9 This is a schematic diagram of a server structure provided in an embodiment of this disclosure. The server 900 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 901 and one or more memories 902. The memories 902 store at least one instruction, which is loaded and executed by the processor 901 to implement the methods provided in the various method embodiments described above. Of course, the server may also have wired or wireless network interfaces, a keyboard, and input / output interfaces for input and output. The server may also include other components for implementing device functions, which will not be elaborated upon here.

[0167] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including instructions that can be executed by a processor in a terminal to complete the method of sending display information in the above embodiments. This computer-readable storage medium may be non-transitory. For example, the computer-readable storage medium may be ROM (read-only memory), RAM (random access memory), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0168] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0169] It should be noted that all information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.), and signals (including but not limited to signals transmitted between user terminals and other devices) involved in this disclosure are authorized by the user or fully authorized by all parties, and the collection, use, and processing of related data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. For example, the "initial search information," "object to be displayed," and "display information" involved in this disclosure were all obtained under full authorization.

[0170] The above description is merely an optional embodiment of this disclosure and is not intended to limit this disclosure. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the protection scope of this disclosure.

Claims

1. A method for sending display information, characterized in that, The method includes: Obtain the initial search information input by the target account and the multiple information units included in the initial search information; Based on the initial search information, candidate display objects are determined from the plurality of objects to be displayed; Based on the display information of the multiple information units and multiple candidate display objects, determine the importance score of each information unit to the display information of each candidate display object; Based on the initial search information, the importance score corresponding to each information unit, and the rewriting model, the rewriting information corresponding to the initial search information is determined; Based on the rewritten information, a display object is determined from the plurality of objects to be displayed, and the display information of the display object is sent to the target account.

2. The method according to claim 1, characterized in that, The importance score includes positive attribution values ​​and / or negative attribution values.

3. The method according to claim 1, characterized in that, The step of determining candidate display objects from the plurality of objects to be displayed based on the initial search information includes: Based on the initial search information, the plurality of objects to be displayed, and the semantic matching model, a first candidate display object corresponding to the initial search information is determined from the plurality of objects to be displayed. Based on the initial search information, the plurality of objects to be displayed, and the keyword matching model, a second candidate object corresponding to the initial search information is determined from the plurality of objects to be displayed. The candidate display object is determined based on the first candidate display object and the second candidate display object.

4. The method according to claim 3, characterized in that, The step of determining the first candidate display object corresponding to the initial search information from among the multiple objects to be displayed, based on the initial search information, the multiple objects to be displayed, and the semantic matching model, includes: Based on the initial search information, the multiple objects to be displayed, and the semantic matching model, determine the semantic matching degree between the initial search information and each object to be displayed; Based on the semantic matching degree, the first candidate display object corresponding to the initial search information is determined among the multiple objects to be displayed; The step of determining the second candidate display object corresponding to the initial search information from among the multiple objects to be displayed, based on the initial search information, the multiple objects to be displayed, and the keyword matching model, includes: Based on the initial search information, the multiple objects to be displayed, and the keyword matching model, determine the keyword matching degree between the initial search information and each object to be displayed; Based on the keyword matching degree, a second candidate display object corresponding to the initial search information is determined from the plurality of objects to be displayed; The step of determining the candidate display object based on the first candidate display object and the second candidate display object includes: Based on the semantic matching degree corresponding to the first candidate display object and the keyword matching degree corresponding to the second candidate display object, the candidate display object is determined from the first candidate display object and the second candidate display object.

5. The method according to claim 4, characterized in that, The step of determining the candidate display object from the first and second candidate display objects based on the semantic matching degree corresponding to the first candidate display object and the keyword matching degree corresponding to the second candidate display object includes: The first ranking information of the first candidate display object is determined based on the semantic matching degree; The second ranking information of the second candidate display object is determined based on the keyword matching degree; Based on the first sorting information and the second sorting information, the third sorting information corresponding to each object to be displayed in the set of display objects composed of the first candidate display objects and the second candidate display objects is determined; Based on the third sorting information, the candidate display objects are determined in the set of display objects.

6. The method according to claim 1, characterized in that, The method further includes: Based on the multiple information units and the multiple attribute information corresponding to the multiple objects to be displayed, the target attribute information corresponding to each information unit is determined from the multiple attribute information; The process of determining the rewrite information corresponding to the initial search information based on the initial search information, the importance score value corresponding to each information unit, and the rewrite model includes: Based on the initial search information, the importance score corresponding to each information unit, the target attribute information corresponding to each information unit, and the rewriting model, the rewriting information corresponding to the initial search information is determined.

7. The method according to claim 6, characterized in that, The method further includes: For each target attribute information, based on the importance score of the information unit corresponding to the target attribute information, the attribute importance score corresponding to each target attribute information is determined; The process of determining the rewriting information corresponding to the initial search information based on the initial search information, the importance score corresponding to each information unit, the target attribute information corresponding to each information unit, and the rewriting model includes: Based on the initial search information, the importance score corresponding to each information unit, the target attribute information corresponding to each information unit, the attribute importance score corresponding to each target attribute information, and the rewriting model, the rewriting information corresponding to the initial search information is determined.

8. The method according to claim 1, characterized in that, The method further includes: The rewritten information and / or the displayed object are verified based on preset verification rules. If the rewritten information does not conform to the preset verification rules, the rewritten information is updated, and the display object is determined based on the updated rewritten information.

9. A device for transmitting display information, characterized in that, The device includes: The acquisition module is used to acquire the initial search information input by the target account and the multiple information units included in the initial search information; The first determining module is used to determine candidate display objects from the plurality of objects to be displayed based on the initial search information. The second determining module is used to determine the importance score of each information unit to the display information of each candidate display object based on the display information of the multiple information units and the multiple candidate display objects. The third determining module is used to determine the rewriting information corresponding to the initial search information based on the initial search information, the importance score value corresponding to each information unit, and the rewriting model. The sending module is used to determine the display object from the plurality of objects to be displayed based on the rewritten information, and send the display information of the display object to the target account.

10. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing at least one instruction, which is loaded and executed by the processor to perform the operation performed by the method of sending display information as described in any one of claims 1 to 8.