Recommendation processing method, device, equipment, program product and storage medium

Abstract

The application provides a recommendation processing method and device, electronic equipment, computer program product and computer readable storage medium. The method comprises: obtaining first attribute information of a to-be-served object, and determining individual scores of multiple indexes of a recommended service object; determining information entropy of each index according to the individual scores of the multiple indexes of the multiple recommended service objects, and determining a weight of each index according to the information entropy of each index and the first attribute information; for each recommended service object, performing weighted processing on the individual scores of the multiple indexes of the recommended service object according to the weight of each index to obtain a comprehensive score of the recommended service object; and performing a first descending sequence sorting on the multiple recommended service objects according to the comprehensive score of each recommended service object, and performing a recommendation operation on the to-be-served object. Through the application, the recommendation accuracy of the service object can be improved.

Description

Technical Field

[0001] This application relates to Internet technology, and more particularly to a recommended processing method, apparatus, electronic device, computer program product, and computer-readable storage medium. Background Technology

[0002] With the development of the Internet, service formats have become more diverse, and the range of recommended service providers to those seeking services has become wider. For example, in the case of advertising services, advertisers, as those seeking services, need to choose from thousands of advertising service agencies (recommended service providers), which makes the selection process difficult and inhibits the efficiency of service selection.

[0003] In related technologies, recommended service objects are ranked and recommended to advertisers based on a certain metric, allowing advertisers to choose according to the ranking results. However, due to the single ranking method, advertisers have a limited range of choices and find it difficult to obtain recommended service objects that match their own preferences. Summary of the Invention

[0004] This application provides a recommendation processing method, apparatus, electronic device, computer program product, and computer-readable storage medium, which can improve the accuracy of recommendations for service recipients.

[0005] The technical solution of this application embodiment is implemented as follows:

[0006] This application provides a recommended processing method, including:

[0007] Obtain the first attribute information of the object to be served, wherein the first attribute information is used to characterize the degree of preference of the object to be served for each indicator of the recommended service object;

[0008] Determine individual scores for each of the aforementioned metrics for each of the recommended service objects;

[0009] Based on the individual scores of multiple metrics for multiple recommended service objects, the information entropy of each metric is determined, and the weight of each metric is determined based on the information entropy of each metric and the first attribute information.

[0010] For each of the recommended service objects, the individual scores of multiple indicators of the recommended service object are weighted according to the weight of each indicator to obtain the comprehensive score of the recommended service object;

[0011] Based on the comprehensive score of each of the recommended service objects, the multiple recommended service objects are sorted in a first descending order, and a recommendation operation is performed on the object to be served based on the first descending order result.

[0012] This application provides a recommended processing apparatus, including:

[0013] The acquisition module is used to acquire the first attribute information of the object to be served, wherein the first attribute information is used to characterize the degree of preference of the object to be served for each indicator of the recommended service object;

[0014] The metrics module is used to determine individual scores for multiple metrics for each of the recommended service objects;

[0015] The indicator module is further configured to determine the information entropy of each indicator based on the individual scores of multiple indicators of multiple recommended service objects, and to determine the weight of each indicator based on the information entropy of each indicator and the first attribute information.

[0016] The indicator module is also used to perform weighted processing on the individual scores of multiple indicators of each recommended service object according to the weight of each indicator, so as to obtain the comprehensive score of the recommended service object.

[0017] The sorting module is used to sort the multiple recommended service objects in a first descending order based on the comprehensive score of each recommended service object, and to perform a recommendation operation on the service object according to the first descending order result.

[0018] In the above scheme, the acquisition module is further configured to: acquire second attribute information of the object to be served before determining the individual scores of multiple indicators for each of the recommended service objects; perform at least one first screening process on multiple candidate recommended service objects based on the second attribute information to obtain a first screening result; acquire third attribute information of the object to be served; query candidate recommended service objects that match the third attribute information of the object to be served from the first screening result to form a target screening result set, wherein the third attribute information is used to characterize the type information of the object to be served; and determine at least one recommended service object that matches the object to be served based on the target screening result set.

[0019] In the above scheme, when the second attribute information includes the industry information of the object to be served, the acquisition module is further configured to: acquire the first filtering object of the first filtering process, and perform the following processing for each candidate recommended service object in the first filtering object: determine the total revenue of the candidate recommended service object, the revenue of the candidate recommended service object corresponding to the industry information, and the revenue corresponding to the industry information; determine a first industry indicator that is negatively correlated with the total revenue of the candidate recommended service object and positively correlated with the revenue of the candidate recommended service object corresponding to the industry information; determine a second industry indicator that is negatively correlated with the revenue of the corresponding industry information and positively correlated with the revenue of the candidate recommended service object corresponding to the industry information; when the first industry indicator is greater than a first threshold and the second industry indicator is greater than a second threshold, retain the candidate recommended service object.

[0020] In the above scheme, when the second attribute information includes at least one level information corresponding to the object to be served, the acquisition module is further configured to: acquire the first filtering object of the first filtering process, and perform the following processing for each candidate recommended service object in the first filtering object: for each level information, determine the income of the candidate recommended service object corresponding to the industry information and the level information, the income corresponding to the industry information and the level information, the income corresponding to the industry information, and the income of the candidate recommended service object corresponding to the industry information; for each level information, determine a level indicator that is positively correlated with the income of the candidate recommended service object corresponding to the industry information and the level information, positively correlated with the income of the candidate recommended service object corresponding to the industry information, negatively correlated with the income of the candidate recommended service object corresponding to the industry information, and negatively correlated with the income of the candidate recommended service object corresponding to the industry information and the level information; when the level indicator obtained for any level information satisfies the level preference condition, the candidate recommended service object is retained.

[0021] In the above scheme, the acquisition module is further configured to: when the type information representation sorts the matched candidate recommendation service objects using a non-independent index sorting method, and the number of matched candidate recommendation service objects does not exceed a number threshold, determine the candidate recommendation service objects in the target screening result set as at least one recommendation service object matching the object to be served; when the type information representation sorts the matched candidate recommendation service objects using a non-independent index sorting method, and the number of matched candidate recommendation service objects does not exceed a number threshold, determine that the direct recommendation condition is met; when the direct recommendation condition is met, perform a recommendation operation on the object to be served based on the candidate recommendation service objects in the target screening result set.

[0022] In the above scheme, the acquisition module is further configured to: when the type information representation sorts the matched candidate recommendation service objects using an independent index ranking method, determine the candidate recommendation service objects in the target screening result set as at least one recommendation service object matching the object to be served; when the type information representation sorts the matched candidate recommendation service objects using an independent index ranking method, determine that the independent index ranking condition is met; when the independent index ranking condition is met, determine the individual scores of the indicators corresponding to the type information of the multiple recommendation service objects, sort the multiple recommendation service objects in a second descending order based on the individual scores, and perform a recommendation operation on the object to be served based on the second descending order ranking result.

[0023] In the above scheme, when the type information representation will sort the matched candidate recommended service objects by a non-independent index sorting method, and the number of matched candidate recommended service objects exceeds the number threshold, the acquisition module is further configured to: acquire the fourth attribute information of the object to be served, and perform at least one second screening process on the target screening result set according to the fourth attribute information to obtain at least one recommended service object that matches the object to be served.

[0024] In the above scheme, the fourth attribute information includes at least one traffic channel information of the object to be served. The acquisition module is further configured to: acquire the second filtering object of the second filtering process, and perform the following processing for each candidate recommended service object in the second filtering object: for each traffic channel information, determine the revenue of the candidate recommended service object corresponding to the industry information and the traffic channel information, the revenue corresponding to the industry information and the traffic channel information, the revenue corresponding to the industry information, and the revenue of the candidate recommended service object corresponding to the industry information; for each traffic channel information, determine the traffic indicators that are positively correlated with the revenue of the candidate recommended service object corresponding to the industry information and the traffic channel information, positively correlated with the revenue of the candidate recommended service object corresponding to the industry information, negatively correlated with the revenue of the candidate recommended service object corresponding to the industry information, and negatively correlated with the revenue of the candidate recommended service object corresponding to the industry information and the traffic channel information; when the traffic indicators obtained for each traffic channel information meet the channel preference conditions, the candidate recommended service object is retained.

[0025] In the above scheme, the acquisition module is further configured to: acquire the second filtering object of the second filtering process, and perform the following processing for each of the candidate recommended service objects in the second filtering object: when the candidate recommended service object meets each of the capability requirement information, the candidate recommended service object is retained.

[0026] In the above scheme, the indicator module is further configured to: determine the indicator value of each indicator for each of the recommended service objects; perform standardization processing on the indicator value of each indicator for each of the recommended service objects to obtain a standardized indicator value; for each indicator, sum the standardized indicator values ​​of multiple recommended service objects corresponding to the indicator to obtain a first summation result, and determine the proportion of the standardized indicator value of each recommended service object corresponding to the indicator in the first summation result as a separate score for each recommended service object corresponding to the indicator.

[0027] In the above scheme, the indicator module is further configured to: perform the following processing for each indicator: perform logarithmic processing on the individual score of each recommended service object corresponding to the indicator, and multiply the logarithmic processing result with the individual score of the corresponding recommended service object corresponding to the indicator; sum the multiplication results of multiple recommended service objects to obtain a second summation result; and determine the information entropy of the indicator that is negatively correlated with the second summation result.

[0028] In the above scheme, the indicator module is further configured to: sort multiple indicators in descending order according to the preference level of each indicator; determine at least one indicator with the highest ranking as a preferred indicator, and determine the remaining indicators as non-preferred indicators; perform the following processing for each preferred indicator: determine the optimized information entropy that is positively correlated with the information entropy of the preferred indicator and a first minimum information entropy, and negatively correlated with a second minimum information entropy; wherein, the first minimum information entropy is the minimum information entropy among the information entropies of multiple indicators, and the second minimum information entropy is the minimum information entropy among the information entropies of at least one preferred indicator; perform the following processing for each non-preferred indicator: determine the information entropy of the non-preferred indicator as the optimized information entropy of the corresponding non-preferred indicator; sum the optimized information entropies of multiple indicators to obtain a total information entropy; and perform the following processing for each indicator: determine the weight of the indicator that is negatively correlated with the optimized information entropy of the indicator and positively correlated with the total information entropy.

[0029] In the above scheme, the sorting module is further configured to: obtain industry information and level information of the object to be served; query multiple historical objects to be served corresponding to the industry information and the level information; determine the number of historical recommended service objects for each historical object to be served; average the number of historical recommended service objects for multiple historical objects to be served to obtain an average number; obtain the recommended service objects that rank first in the first descending sorting result and meet the average number; and perform a recommendation operation on the object to be served based on the recommended service objects that meet the average number.

[0030] This application provides an electronic device, including:

[0031] Memory, used to store executable instructions;

[0032] The processor, when executing executable instructions stored in the memory, implements the recommended processing method provided in the embodiments of this application.

[0033] This application provides a computer-readable storage medium storing executable instructions for implementing the recommended processing method provided in this application when executed by a processor.

[0034] This application provides a computer program product, including a computer program or instructions, which, when executed by a processor, implement the recommended processing method provided in this application.

[0035] The embodiments of this application have the following beneficial effects:

[0036] This application can comprehensively consider the personalized needs of the target users' indicator preferences and the objective indicator scores of the recommended service users. This can ensure that the recommended service users are well matched with the actual situation and needs of the target users, thereby improving the accuracy and efficiency of the recommendation. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of the architecture of the recommendation processing system provided in the embodiments of this application;

[0038] Figure 2 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application;

[0039] Figures 3A-3C This is a flowchart illustrating the recommended processing method provided in the embodiments of this application;

[0040] Figure 4 This is a schematic diagram of the interaction flow of the recommended processing method provided in the embodiments of this application;

[0041] Figure 5 This is a schematic diagram of the recommended processing method provided in the embodiments of this application;

[0042] Figure 6 This is a schematic diagram of the screening process of the recommended processing method provided in the embodiments of this application. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0044] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0045] In the following description, the terms "first, second, third" are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0046] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0047] In the implementation of this application, the collection and processing of relevant data should strictly comply with the requirements of relevant laws and regulations, obtain the informed consent or separate consent of the personal information subject, and carry out subsequent data use and processing within the scope of laws and regulations and the authorization of the personal information subject.

[0048] Before providing a further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained, and the nouns and terms involved in the embodiments of this application shall be interpreted as follows.

[0049] 1) Consumption: The amount spent by the recommended service recipient when providing services to the service recipient, such as the amount spent by the advertiser to place an advertisement.

[0050] 2) Entropy Weight Method: The basic idea of ​​the entropy weight method is to determine the objective weight based on the magnitude of the variability of the indicator. Generally speaking, the smaller the information entropy of an indicator, the greater the degree of variation of the indicator value, the more information it provides, and the greater its role in the comprehensive evaluation, thus its weight is also greater. Conversely, the larger the information entropy of an indicator, the smaller the degree of variation of the indicator value, the less information it provides, and the smaller its role in the comprehensive evaluation, thus its weight is also smaller.

[0051] 3) Successful ad campaigns: Ads with a conversion rate of more than 30 within 7 days.

[0052] 4) Conversions: Before launching an ad campaign, advertisers select the goals they hope the ad will achieve, such as clicks, installations, activations, registrations, etc. The number of these goals achieved is the conversion rate.

[0053] 5) Successful scaling accounts: This account contains ads that have successfully scaled up. For example, when recommending an advertiser to a client in the financial industry, this is the number of accounts that the service provider has successfully scaled up in the financial industry.

[0054] 6) Cost Achievement: Cost is the ratio of the actual bid to the target bid. When the cost is between 0.8 and 1.2, it means that the cost has been achieved. The actual bid is the real cost of the ad to achieve a conversion, and the target bid is the cost that the client is willing to pay for each conversion before launching the ad.

[0055] 7) Increased volume advertising: Ads in which the advertiser's recharge amount increases by more than 20% compared to the previous period within a certain time period.

[0056] In related technologies, the selection of recommended service providers by the service recipients relies more on the service provider's size in the current industry and its own marketing efforts, lacking a professional recommendation system to support the decision-making process. When implementing the embodiments of this application, the applicant discovered the following problems with the methods of selecting recommended service providers in related technologies:

[0057] First, the selection of recommended service targets is often based solely on the amount of money they have invested in the industry. This ignores the overall capabilities of the recommended service targets. Recommended service targets that have invested a large amount of money in the industry may not have outstanding overall capabilities, resulting in poor advertising performance.

[0058] Secondly, clients have many personalized requirements for the capabilities of recommended service providers, such as video production capabilities, technical service capabilities, and mini-program development capabilities. Moreover, the promotion and marketing of recommended service providers may exaggerate their advertising effects and operational capabilities, making it difficult for clients to distinguish the capability differences between different recommended service providers. If a poor choice is made, it will also affect the advertising effectiveness.

[0059] Third, if the target customer is small in size, the recommended service providers with a large market share in the industry will not have the resources to provide services, making it difficult for small-sized target customers to choose recommended service providers.

[0060] To address the aforementioned technical problems, embodiments of this application provide a recommendation processing method, apparatus, electronic device, computer program product, and computer-readable storage medium, which can comprehensively consider the personalized needs of the target user's indicator preferences and the objective indicator scores of the recommended service user, and execute recommendations based on the comprehensive consideration results.

[0061] The recommended processing method provided in this application embodiment can be implemented by various electronic devices, such as by a terminal or server alone, or by a terminal and server working together.

[0062] See Figure 1 , Figure 1 This is a schematic diagram of the architecture of the recommendation processing system provided in the embodiments of this application. The terminal 400 is connected to the server 200 through the network 300. The network 300 can be a wide area network or a local area network, or a combination of the two.

[0063] In some embodiments, the function of the recommendation processing system is implemented using server 200 and terminal 400. Taking the advertiser as the target of service and the service provider as the recommended service target as an example, when the target of service uses terminal 400, in response to the target of service receiving the trigger operation of the recommendation function control, terminal 400 displays the attribute information input interface. In response to the target of service receiving the attribute information input by the target of service, terminal 400 sends the attribute information to server 200. Server 200 determines the comprehensive score of the recommended service target based on the attribute information and the individual scores of multiple indicators for each recommended service target. Based on the comprehensive score of each recommended service target, multiple recommended service targets are sorted in a first descending order. Server 200 returns the recommended service targets in the first descending order to terminal 400, and the recommended service targets are displayed on terminal 400 according to the first descending order.

[0064] In other embodiments, when the recommendation processing method provided in this application is implemented by a terminal alone, taking the advertiser as an example, during the use of the terminal 400 by the advertiser, in response to the terminal 400 receiving the advertiser's trigger operation on the recommendation function control, the terminal 400 displays an attribute information input interface. In response to the terminal 400 receiving the attribute information input by the advertiser, the terminal 400 determines the comprehensive score of the recommended service object based on the attribute information and the individual scores of multiple indicators for each recommended service object. Based on the comprehensive score of each recommended service object, the multiple recommended service objects are sorted in a first descending order, and the recommended service objects are displayed on the terminal 400 according to the first descending order sorting result.

[0065] In some embodiments, server 200 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. Terminal 400 can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, smart TV, smart in-vehicle device, etc. Terminal 400 may have a client application, such as a video client, browser client, information streaming client, image capture client, etc., but is not limited to these. The terminal and server can be directly or indirectly connected via wired or wireless communication, which is not limited in this embodiment.

[0066] In some embodiments, the terminal or server can implement the recommended processing method provided in the embodiments of this application by running a computer program. For example, the computer program can be a native program or software module in an operating system; it can be a native application (APP), that is, a program that needs to be installed in the operating system to run, such as an advertising app; it can also be a mini-program, that is, a program that only needs to be downloaded into a browser environment to run; or it can be a mini-program that can be embedded in any APP. In short, the above-mentioned computer program can be any form of application, module or plugin.

[0067] Next, the structure of the electronic device for implementing the recommended processing method provided in the embodiments of this application will be described. As before, the electronic device provided in the embodiments of this application may be... Figure 1 Server 200 or Terminal 400. See also Figure 2 , Figure 2 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device is described using terminal 400 as an example. Figure 2 The terminal 400 shown includes at least one processor 410, a memory 450, and at least one network interface 420. The various components in the terminal 400 are coupled together via a bus system 440. It is understood that the bus system 440 is used to implement communication between these components. In addition to a data bus, the bus system 440 also includes a power bus, a control bus, and a status signal bus. However, for clarity, ... Figure 2 The general labeled all buses as Bus System 440.

[0068] Processor 410 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0069] User interface 430 includes one or more output devices 431 that enable the presentation of media content, including one or more speakers and / or one or more visual displays. User interface 430 also includes one or more input devices 432, including user interface components that facilitate user input, such as a keyboard, mouse, microphone, touch screen display, camera, other input buttons and controls.

[0070] The memory 450 may be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state storage, hard disk drives, optical disk drives, etc. The memory 450 may optionally include one or more storage devices physically located away from the processor 410.

[0071] The memory 450 may include volatile memory or non-volatile memory, or both. The non-volatile memory may be read-only memory (ROM), and the volatile memory may be random access memory (RAM). The memory 450 described in this application embodiment is intended to include any suitable type of memory.

[0072] In some embodiments, memory 450 is capable of storing data to support various operations, examples of which include programs, modules, and data structures or subsets or supersets thereof, as illustrated below.

[0073] Operating system 451 includes system programs for handling various basic system services and performing hardware-related tasks, such as the framework layer, core library layer, driver layer, etc., for implementing various basic business functions and handling hardware-specific tasks.

[0074] The network communication module 452 is used to reach other electronic devices via one or more (wired or wireless) network interfaces 420, exemplary network interfaces 420 including: Bluetooth, WiFi, and Universal Serial Bus (USB), etc.

[0075] Presentation module 453 is configured to enable the presentation of information (e.g., a user interface for operating peripheral devices and displaying content and information) via one or more output devices 431 (e.g., a display screen, a speaker, etc.) associated with user interface 430.

[0076] The input processing module 454 is used to detect and translate one or more user inputs or interactions from one or more input devices 432.

[0077] In some embodiments, the recommended processing apparatus provided in this application can be implemented in software. Figure 2 A recommendation processing device 455 stored in memory 450 is shown. It may be software in the form of programs and plug-ins, including the following software modules: acquisition module 4551, index module 4552, and sorting module 4553. These modules are logical and can therefore be arbitrarily combined or further divided according to the functions they implement. The functions of each module will be described below.

[0078] The recommended processing method provided in this application will be described in conjunction with exemplary applications and implementations of the terminal provided in the embodiments of this application.

[0079] See Figure 3A , Figure 3A This is a flowchart illustrating the recommended processing method provided in the embodiments of this application, which will be combined with... Figure 3A Steps 101-105 shown are explained.

[0080] In step 101, the first attribute information of the object to be served is obtained.

[0081] As an example, the first attribute information is used to characterize the degree of preference of the service recipient for each metric of the recommended service recipient.

[0082] As an example, the "service recipient" is the object receiving the service. For instance, relative to the advertising service provider and the advertiser, the advertiser is the service recipient, and the advertising service provider is the recommended service recipient. In response to the service recipient inputting first attribute information in the form of a question-and-answer session through the human-computer interaction interface, the service recipient's first attribute information is obtained. The first attribute information can also be obtained from the service recipient's historical data. The first attribute information includes the service recipient's preference for each indicator of the recommended service recipient. For example, there are three indicators A, B, and C used to measure the recommended service recipient, but the service recipient values ​​indicator A more. For example, if the service recipient selects indicator A, then indicator A is taken as the preferred indicator. For example, if the service recipient scores the three indicators, and indicator A scores 50 points, indicator B scores 70 points, and indicator C scores 90 points, then the indicator with a score greater than the score threshold is taken as the service recipient's preferred indicator.

[0083] In step 102, individual scores for multiple metrics are determined for each recommended service object.

[0084] In some embodiments, see Figure 3B , Figure 3B This is a flowchart illustrating the recommendation processing method provided in an embodiment of this application. In step 102, the individual scores of multiple indicators for each recommended service object can be determined through... Figure 3B Steps 1021-1023 shown are implemented.

[0085] In step 1021, the indicator value for each indicator of each recommended service object is determined.

[0086] As an example, taking an advertising service agency as the target of the recommended service, the metrics include at least one of the following: scaling loss, scaling speed, conversion efficiency, cost achievement rate, and scaling capability. The process involves obtaining industry information about the target service provider and performing at least one of the following: determining the initial ad spend for new ads corresponding to the target service provider's industry information, and the first total ad spend for new ads corresponding to the target service provider's industry information; determining the scaling loss metric value that is positively correlated with the initial ad spend and negatively correlated with the first total spend; determining the number of accounts that successfully scaled up the target service provider's industry information, and the total number of new accounts corresponding to the target service provider's industry information; and determining the value that is positively correlated with the number of accounts that successfully scaled up the target service provider's industry information and negatively correlated with the total number of new accounts. The metrics include: the speed of initialization; the conversion rate and exposure of the industry information corresponding to the recommended service target; the conversion efficiency metric, which is positively correlated with conversion rate and negatively correlated with exposure; the cost-effectiveness ad spend and the second total spend of all ads for the industry information corresponding to the recommended service target; the cost-effectiveness rate metric, which is positively correlated with cost-effectiveness ad spend and negatively correlated with the second total spend; and the spending and recharge amounts of the expansion ads for the industry information corresponding to the recommended service target, which are positively correlated with spending and negatively correlated with recharge amounts.

[0087] In step 1022, the index values ​​of each index for each recommended service object are standardized to obtain standardized index values.

[0088] As an example, the index values ​​of each indicator for each recommended service object are first standardized using formula (1):

[0089] (1);

[0090] in, It is the standardized metric value of the j-th metric for the i-th recommended service object. It is the value of the j-th metric for the i-th recommended service object. It is the minimum value among multiple metrics for the i-th recommended service object. It is the maximum value among multiple metrics of the i-th recommended service object.

[0091] In step 1023, for each indicator, the standardized indicator values ​​of multiple recommended service objects are summed to obtain a first summation result, and the proportion of the standardized indicator value of each recommended service object to the first summation result is determined as the individual score of each recommended service object's corresponding indicator.

[0092] As an example, the proportion of the index value of the j-th indicator of the i-th recommended service object to the first summation result of the index values ​​of all recommended service objects for the j-th indicator is calculated using formula (2):

[0093] (2);

[0094] in, It is the individual score of the j-th metric for the i-th recommended service object. is the standardized index value of the j-th metric for the i-th recommended service object, and n is the number of recommended service objects.

[0095] In step 103, the information entropy of each indicator is determined based on the individual scores of multiple indicators for multiple recommended service objects, and the weight of each indicator is determined based on the information entropy of each indicator and the first attribute information.

[0096] In some embodiments, determining the information entropy of each indicator in step 103 based on the individual scores of multiple indicators for multiple recommended service objects can be achieved through the following technical solution: For each indicator, perform the following processing: perform logarithmic processing on the individual score of the indicator corresponding to each recommended service object, and multiply the logarithmic processing result with the individual score of the corresponding indicator of the recommended service object; sum the results of the multiplication of multiple recommended service objects to obtain a second summation result; determine the information entropy of the indicator negatively correlated with the second summation result.

[0097] As an example, the information entropy of each indicator is calculated using formula (3):

[0098] (3);

[0099] in, It is the information entropy of the j-th index. It is the individual score of the j-th metric for the i-th recommended service object. It is the logarithmic result of the individual score of the i-th recommended service object corresponding to the j-th indicator. This is the second summation result, where n is the number of recommended service recipients.

[0100] In some embodiments, determining the weight of each indicator based on its information entropy and the first attribute information in step 103 can be achieved through the following technical solution: sorting multiple indicators in descending order according to the degree of preference for each indicator; identifying at least one indicator ranked first as a preferred indicator and the remaining indicators as non-preferred indicators; performing the following processing for each preferred indicator: determining the optimized information entropy of the corresponding preferred indicator that is positively correlated with the information entropy of the preferred indicator and the first minimum information entropy, and negatively correlated with the second minimum information entropy; wherein the first minimum information entropy is the minimum information entropy among the information entropies of multiple indicators, and the second minimum information entropy is the minimum information entropy among the information entropies of at least one preferred indicator; performing the following processing for each non-preferred indicator: determining the information entropy of the non-preferred indicator as the optimized information entropy of the corresponding non-preferred indicator; summing the optimized information entropies of multiple indicators to obtain the total information entropy; performing the following processing for each indicator: determining the weight of the indicator that is negatively correlated with the optimized information entropy of the indicator and positively correlated with the total information entropy.

[0101] As an example, based on business experience, the information entropy value is optimized. When there are 5 indicators, taking the first and second indicators as preferred indicators and the third, fourth, and fifth indicators as unpreferred indicators, the weights of the first and second indicators are increased, even to the point that the weight of one of the first and second indicators is the maximum weight of the 5 indicators. See formulas (4) and (5).

[0102] (4);

[0103] (5);

[0104] in, It is the optimized information entropy of the j-th index. It refers to the information entropy of the first indicator and the information entropy of the second indicator. It is the minimum value between the information entropy of the first indicator and the information entropy of the second indicator. These are the information entropy of the third, fourth, and fifth indicators. It is the minimum value among the information entropy of the third, fourth, and fifth indicators.

[0105] As an example, the weight of each indicator is calculated using formula (6):

[0106] (6);

[0107] in, It is the weight of the j-th indicator. It is the optimized information entropy of the j-th index, where k is a constant. It is the sum of the optimized information entropy of all indicators.

[0108] In step 104, for each recommended service object, the individual scores of multiple indicators of the recommended service object are weighted according to the weight of each indicator to obtain the comprehensive score of the recommended service object.

[0109] As an example, the comprehensive score of the indicator is calculated using formula (7):

[0110] (7);

[0111] in, It is the overall score of the i-th recommended service recipient. It is the individual score of the j-th metric for the i-th recommended service object. This represents the weight of the j-th indicator. Finally, a comprehensive score for each recommended service object is obtained, and the recommended service objects are ranked in descending order based on this comprehensive score.

[0112] In step 105, based on the comprehensive score of each recommended service object, the multiple recommended service objects are sorted in a first descending order, and a recommendation operation is performed on the service object according to the first descending order result.

[0113] In some embodiments, see Figure 3C , Figure 3C This is a flowchart illustrating the recommendation processing method provided in this application embodiment. In step 105, the recommendation operation is performed on the objects to be served based on the first descending sort result. Figure 3C Steps 1051-1056 shown are implemented.

[0114] In step 1051, the industry information and grade information of the target customer are obtained.

[0115] In step 1052, multiple historical pending service objects with corresponding industry information and grade information are queried.

[0116] In step 1053, the number of historical recommended service objects for each historical pending service object is determined.

[0117] In step 1054, the number of historical recommended service objects for multiple historical pending service objects is averaged to obtain the average number.

[0118] In step 1055, the recommended service objects that meet the average number of sorted items at the top of the first descending sort result are obtained.

[0119] In step 1056, a recommendation operation is performed on the target users based on the average number of recommended service users.

[0120] As an example, the top M recommended service objects in the descending score results are truncated, and the client is advised to select N recommended service objects. M is calculated as follows: obtain historical clients in the same industry and level as the client, obtain the average number of historical recommended service objects used by each historical client, which can be the average number of historical recommended service objects selected in a single selection, or the average number of historical recommended service objects selected over a period of time.

[0121] In some embodiments, when performing a recommendation operation on a client based on an average number of recommended service providers, the client may be prompted to select N recommended service providers from M recommended service providers. N is the number of historical recommended service providers that cover 80% of the service volume used by historical clients in the same industry and level as the client. This is because among the historical recommended service providers selected by the client, some historical recommended service providers account for a large proportion of the client's service volume, while others only receive a small share. Therefore, it is recommended that the client select only the number of historical recommended service providers that cover 80% of the consumption. If there are too many service providers, it will lead to excessive communication costs for the client.

[0122] In some embodiments, the recommended service objects are obtained by filtering from candidate recommended service objects. Before determining the individual scores of multiple indicators for each recommended service object, the second attribute information of the object to be served is obtained, and the multiple candidate recommended service objects are subjected to at least one first filtering process based on the second attribute information to obtain a first filtering result; the third attribute information of the object to be served is obtained, and candidate recommended service objects that match the third attribute information of the object to be served are queried from the first filtering result to form a target filtering result set, wherein the third attribute information is used to characterize the type information of the object to be served; based on the target filtering result set, at least one recommended service object that matches the object to be served is determined.

[0123] As an example, the recommended service objects are obtained by filtering from the candidate recommended service objects. The filtering process can be done as follows: Figure 5The filtering process is performed sequentially, for example, first by industry preference filter, then by customer size filter, and then by type information to determine whether to proceed to the next step. If further filtering is required, the process continues by using traffic preference filter and support capability filter. The final filtered results are used as recommended service objects to continue executing steps 101 to 105. Alternatively, all candidate recommended service objects can be used directly as recommended service objects without filtering, and steps 101 to 105 can continue to be executed.

[0124] As an example, for the technical solutions of steps 101 to 105, the solution to execute steps 101 to 105 can be determined when the index weighted ranking conditions are met. For example, the index weighted ranking conditions are that the number of candidate recommended service objects exceeds the number threshold, and the service object needs to comprehensively consider multiple indicators for the recommended service object. For example, the service object has checked the control for multi-indicator composite ranking.

[0125] In some embodiments, when the second attribute information includes industry information of the object to be served, any first screening process can be implemented by the following technical solution: obtaining the first screening object of the first screening process, and performing the following processing for each candidate recommended service object in the first screening object: determining the total revenue of the candidate recommended service object, the revenue of the industry information corresponding to the candidate recommended service object, and the revenue of the corresponding industry information; determining a first industry indicator that is negatively correlated with the total revenue of the candidate recommended service object and positively correlated with the revenue of the industry information corresponding to the candidate recommended service object; determining a second industry indicator that is negatively correlated with the revenue of the corresponding industry information and positively correlated with the revenue of the industry information corresponding to the candidate recommended service object; when the first industry indicator is greater than a first threshold and the second industry indicator is greater than a second threshold, retaining the candidate recommended service object.

[0126] As an example, the industry preference filter mainly evaluates the depth of a candidate service provider's presence in the industry. The first industry indicator is the ratio of the candidate service provider's revenue in the client's industry to the candidate service provider's total revenue. The second industry indicator is the ratio of the candidate service provider's revenue in the client's industry to the revenue in the current industry. When the values ​​of both indicators are greater than their respective medians, the candidate service provider proceeds to the next step of the calculation; otherwise, it is directly removed from the recommendation range.

[0127] In some embodiments, when the second attribute information includes at least one level information corresponding to the object to be served, any first screening process can be implemented by the following technical solution: obtaining the first screening object of the first screening process, and performing the following processing for each candidate recommended service object in the first screening object: for each level information, determining the revenue of the industry information and level information corresponding to the candidate recommended service object, the revenue of the industry information and level information corresponding to the corresponding industry information, the revenue of the industry information corresponding to the candidate recommended service object, and the revenue of the industry information corresponding to the candidate recommended service object; for each level information, determining the level index that is positively correlated with the revenue of the industry information and level information corresponding to the candidate recommended service object, positively correlated with the revenue of the industry information corresponding to the candidate recommended service object, negatively correlated with the revenue of the industry information and level information corresponding to the candidate recommended service object, and negatively correlated with the revenue of the industry information and level information corresponding to the candidate recommended service object; when the level index obtained for any level information satisfies the level preference condition, the candidate recommended service object is retained.

[0128] As an example, the primary function of the service target size preference filter is to evaluate whether candidate recommended service targets can handle the size of the service target. See [link to relevant documentation]. Figure 6 , Figure 6 This is a schematic diagram of the screening process of the recommendation processing method provided in this application embodiment. Investments greater than 500 million are classified as S-level candidates, investments between 500 million and 100 million as A-level candidates, investments between 100 million and 10 million as B-level candidates, investments between 10 million and 1 million as C-level candidates, and investments less than 1 million as D-level candidates. The candidate's level is determined based on their budget. A level index is calculated for each candidate. The numerator of the level index is the ratio of the candidate's income in their industry and level to the income in their industry. The denominator is the ratio of the income in their industry and level to the income in their industry. Calculations are performed for multiple levels, including calculating the candidate's current level, the result of decreasing the candidate's level by one level, and the result of increasing the candidate's level by one level. If any result is greater than the corresponding median, the candidate proceeds to the next calculation step; otherwise, it is removed from the recommendation range.

[0129] For example, if the target object is level A, then we need to substitute it into the calculation to get three results: level S, level A, and level B. If the target object is level S, then we need to calculate two results: level S and level A. If the target object is level D, then we need to calculate two results: level C and level D. The advantage of calculating three adjacent levels is that it ensures that the recommended candidate service objects can cover the scale of the target object, and that there are enough candidate recommended service objects for the target object to choose from.

[0130] As an example, at least one first filtering process is performed sequentially, and each first filtering object is the filtering result of the previous first filtering process. If the current first filtering process is the first first filtering process, then the first filtering object of the current first filtering process is the original candidate recommendation service object.

[0131] In some embodiments, when the type information representation sorts the matched candidate recommended service objects using a non-independent index ranking method, and the number of matched candidate recommended service objects does not exceed a number threshold, determining at least one recommended service object matching the object to be served based on the target screening result set can be achieved through the following technical solution: determining the candidate recommended service objects in the target screening result set as at least one recommended service object matching the object to be served; when the type information representation sorts the matched candidate recommended service objects using a non-independent index ranking method, and the number of matched candidate recommended service objects does not exceed a number threshold, determining that the direct recommendation condition is met; when the direct recommendation condition is met, performing a recommendation operation on the object to be served based on the candidate recommended service objects in the target screening result set.

[0132] In some embodiments, when the type information representation sorts the matched candidate recommended service objects using an independent index ranking method, determining at least one recommended service object matching the object to be served based on the target screening result set can be achieved through the following technical solution: determining the candidate recommended service objects in the target screening result set as at least one recommended service object matching the object to be served; when the type information representation sorts the matched candidate recommended service objects using an independent index ranking method, determining that the independent index ranking condition is met; when the independent index ranking condition is met, determining the individual scores of the corresponding type information indicators of multiple recommended service objects, ranking the multiple recommended service objects in a second descending order based on the individual scores, and performing a recommendation operation on the object to be served based on the second descending order ranking result.

[0133] As an example, after at least one initial screening process, a query is performed based on type information. When the type information representation uses a non-independent index ranking method to sort the matched candidate recommended service objects, and the number of matched candidate recommended service objects does not exceed a threshold, for example, see [link to relevant documentation]. Figure 5If the service recipient type is advance payment, and the main demand of the service recipient is not various operational indicators, then the candidate recommended service recipients with strong financial capabilities are directly selected as recommended service recipients. These recommended service recipients are then sorted in a second descending order based on financial capabilities, and then recommendations are made based on the results of the second descending order. If the service recipient type is operation-related, and the service recipient is a regional service recipient, since the number of candidate recommended service recipients in that region is limited, then the candidate recommended service recipients in that region are directly selected as recommended service recipients, and the corresponding recommendation operations for these recommended service recipients are executed for the service recipient.

[0134] In some embodiments, when the type information representation will sort the matched candidate recommended service objects using a non-independent index sorting method, and the number of matched candidate recommended service objects exceeds a threshold, at least one recommended service object matching the object to be served can be determined based on the target filtering result set. This can be achieved through the following technical solution: obtaining the fourth attribute information of the object to be served, and performing at least one second filtering process on the target filtering result set based on the fourth attribute information to obtain at least one recommended service object matching the object to be served.

[0135] In some embodiments, the fourth attribute information includes at least one traffic channel information of the object to be served. Any second filtering process can be implemented by the following technical solution: obtaining the second filtering object of the second filtering process, and performing the following processing for each candidate recommended service object in the second filtering object: for each traffic channel information, determining the revenue of the corresponding industry information and traffic channel information of the candidate recommended service object, the revenue of the corresponding industry information and traffic channel information, the revenue of the corresponding industry information, and the revenue of the corresponding industry information of the candidate recommended service object; for each traffic channel information, determining the traffic indicators that are positively correlated with the revenue of the corresponding industry information and traffic channel information of the candidate recommended service object, positively correlated with the revenue of the corresponding industry information, negatively correlated with the revenue of the corresponding industry information and traffic channel information of the candidate recommended service object, and negatively correlated with the revenue of the corresponding industry information and traffic channel information; when the traffic indicators obtained for each traffic channel information meet the channel preference conditions, the candidate recommended service object is retained.

[0136] As an example, in the traffic preference filter, if a customer selects preferred traffic types (multiple selections are allowed), the traffic preference filter is triggered to perform filtering. Ad traffic is divided into four categories: social media traffic, public platform traffic, video news traffic, affiliate traffic, etc. Traffic metrics are calculated for each of the four types of traffic. Taking social media traffic as an example, the numerator of the traffic metric is the ratio of the service provider's revenue from advertising through social media traffic in the customer's industry to the service provider's total revenue from advertising through social media traffic in the customer's industry. The denominator of the traffic metric is the ratio of the revenue from advertising through social media traffic in the customer's industry to the total revenue from advertising through advertising through social media traffic in the customer's industry. Based on the customer's selected preferred traffic, if the service provider has at least one traffic type with a value greater than the median, the calculation proceeds to the next step; otherwise, the service provider is removed from the recommendation pool.

[0137] In some embodiments, the fourth attribute information includes at least one capability requirement information of the object to be served. Any second screening process can be implemented by the following technical solution: obtaining the second screening object of the second screening process, and performing the following processing for each candidate recommended service object in the second screening object: when the candidate recommended service object meets each capability requirement information, the candidate recommended service object is retained.

[0138] As an example, the capability filter allows clients to make multiple selections to meet their advertising needs. Specifically, it offers three capability filters: technical service provider capability (whether the provider can provide technical support for the market's application programming interface), mini-program creation capability (whether the provider can create mini-programs), and video material production capability (whether the provider provides video material production services). If the client selects mini-program creation capability, service providers without this capability will be removed from the recommendation pool. Otherwise, the process proceeds to the next step. This concludes the filtering module, resulting in a shortlist of recommended service providers. These recommended service providers are then input into the sorting module.

[0139] As an example, at least one second filtering process is performed sequentially, and each second filtering object is the filtering result of the previous second filtering process. If the current second filtering process is the first second filtering process, then the second filtering object of the current second filtering process is the original candidate recommendation service object.

[0140] This application embodiment can screen service providers based on the advertiser's industry and scale, comprehensively considering the advertiser's personalized needs such as traffic preferences and operational capability preferences with the service providers' objective operational capabilities and financial resources. Finally, it recommends candidate service providers to the advertiser based on the advertiser's scale. This ensures that the recommended service providers can well match the advertiser's actual situation and needs in terms of industry, scale, and various capabilities. Thus, this invention can help advertisers find the best matching service providers and improve the effectiveness of their advertising campaigns.

[0141] The following will describe an exemplary application of the embodiments of this application in a real-world application scenario.

[0142] In some embodiments, the function of the recommendation processing system is implemented based on a server and a terminal. Taking an advertiser as the target of service and a service provider as the recommended service target as an example, when the target of service uses the terminal, in response to the terminal receiving a trigger operation from the target of service for the recommendation function control, the terminal displays an attribute information input interface. In response to the terminal receiving the attribute information input by the target of service, the terminal sends the attribute information to the server. The server determines the comprehensive score of the recommended service target based on the attribute information and the individual scores of multiple indicators for each recommended service target. Based on the comprehensive score of each recommended service target, the multiple recommended service targets are sorted in a first descending order. The server returns the recommended service targets in the first descending order to the terminal, and the recommended service targets are displayed on the terminal according to the first descending order.

[0143] In some embodiments, see Figure 4 , Figure 4 This is a schematic diagram of the interactive process of the recommendation processing method provided in this application embodiment. When the customer is an advertiser and the customer creates an advertising account, the advertising system prompts the service provider recommendation function. When the customer clicks to enter the service provider recommendation system, they need to answer four mandatory questions and three optional questions. The service provider recommendation system will automatically calculate the service providers that best match the customer based on the information (attribute information) entered by the customer, display a list of M service providers, and prompt the customer to select N service providers from them to be responsible for the customer's advertising services through copywriting. M and N are values ​​calculated based on the customer's size. In this way, the customer can consult and contact the displayed service providers through the information.

[0144] In some embodiments, see Figure 5 The service provider recommendation system is divided into three main modules: a filtering module, a sorting module, and a ranking truncation module. In the filtering module, based on the answers (attribute information) to the four mandatory questions and two optional questions entered by the customer, a list of service providers that best match the customer's current situation and preferences is filtered out. The data time range for the calculation can be the most recent year.

[0145] In some embodiments, the filtering module includes an industry preference filter, which is mainly used to evaluate the service provider's depth of involvement in the industry. The first industry indicator is the service provider's current industry as a percentage of its consumption, and the second industry indicator is the ratio of the service provider's revenue in the customer's industry to the service provider's total revenue. The third industry indicator is the service provider's current industry as a percentage of the total consumption in the industry, and the fourth industry indicator is the service provider's revenue in the customer's industry to its current industry revenue. When the values ​​of both indicators are greater than their respective medians, the service provider proceeds to the next step of the calculation; otherwise, it is directly removed from the recommendation range.

[0146] In some embodiments, the filtering module may further include a customer volume preference filter, whose main function is to evaluate whether the service provider can handle the customer volume. See [link to relevant documentation]. Figure 6 Customers with investments exceeding 500 million are classified as S-level customers, those between 500 million and 100 million as A-level customers, those between 100 million and 10 million as B-level customers, those between 10 million and 1 million as C-level customers, and those below 1 million as D-level customers. Customer levels are determined based on their budgets. A level indicator is calculated for each service provider. The numerator of the level indicator is the ratio of the service provider's revenue in the customer's industry and at the customer's level to the service provider's revenue in the customer's industry. The denominator is the ratio of the customer's revenue in the customer's industry and at the customer's level to the customer's industry revenue. Calculations are performed for multiple levels, including calculating the customer's current level, the result of downgrading the customer's level by one level, and the result of upgrading the customer's level by one level. If any result is greater than the corresponding median, the service provider proceeds to the next step of the calculation; otherwise, it is removed from the recommendation pool.

[0147] For example, if the customer is level A, then we need to substitute the values ​​to calculate level S, level A, and level B to get three results. If the customer level is level S, then we need to calculate level S and level A. If the customer level is level D, then we need to calculate level C and level D. The advantage of calculating three adjacent levels is that it ensures that the recommended service provider can handle the customer volume and that there are enough service providers for the customer to choose from.

[0148] In some embodiments, if the customer type is advance payment, the customer's main need is not operation, so it is sufficient to directly recommend a service provider with strong financial capabilities. If the customer is a regional customer, it is sufficient to directly recommend a contract signing service provider in that region.

[0149] In some embodiments, the filtering module may further include a traffic preference filter. If a customer selects preferred traffic (multiple selections are possible), the traffic preference filter is triggered to perform filtering. Advertising traffic is divided into four categories: social media traffic, public platform traffic, video news traffic, affiliate traffic, etc. Traffic metrics are calculated for each of the four types of traffic. Taking social media traffic as an example, the numerator of the traffic metric is the ratio of the service provider's revenue from advertising through social media traffic in the customer's industry to the service provider's total revenue from advertising through social media traffic in the customer's industry. The denominator of the traffic metric is the ratio of the revenue from advertising through social media traffic in the customer's industry to the total revenue from advertising through advertising through social media traffic in the customer's industry. Based on the customer's selected preferred traffic, if the service provider has at least one traffic type with a value greater than the median, the calculation proceeds to the next step; otherwise, the service provider is directly removed from the recommendation range.

[0150] In some embodiments, the filtering module may also include a capability filter. To meet customer deployment needs, there are no default requirements, but customers can be provided with multiple selection options. Specifically, three capability filters are provided: technical service provider capability (whether the service provider can provide technical support for the market application programming interface), mini-program creation capability (whether the service provider can create mini-programs), and video material production capability (whether the service provider provides video material production services). If the customer selects mini-program creation capability, service providers that do not have mini-program creation capability will be directly removed from the recommendation range. Otherwise, the subsequent judgment process will proceed.

[0151] The filtering module has now ended, and the list of shortlisted service providers, i.e., recommended service recipients, is now obtained. These recommended service recipients are then input into the sorting module.

[0152] In some embodiments, the ranking module comprehensively ranks the service providers' operational capabilities based on customer metric preferences to obtain a comprehensive operational score ranking of the shortlisted service providers. The operational capability items (metrics) are: scaling-up cost (corresponding to the metric value of scaling-up cost), scaling-up speed (corresponding to the metric value of scaling-up speed), conversion efficiency (corresponding to the metric value of conversion efficiency), cost achievement rate (corresponding to the metric value of cost achievement rate), and scaling capability (corresponding to the metric value of scaling capability).

[0153] In some embodiments, the scaling-up cost is used to evaluate whether the service provider wastes a lot of money from the client when advertising scales up. The scaling-up cost score is the ratio of the service provider's scaling-up advertising cost when placing new ads in the client's industry to the service provider's total cost when placing new ads in the client's industry.

[0154] In some embodiments, scaling speed is used to evaluate a service provider’s scaling capability. The scaling speed score is the ratio of the number of accounts successfully scaled by the service provider in the client’s industry to the total number of new advertising accounts the service provider has in the client’s industry.

[0155] In some embodiments, conversion efficiency is used to evaluate a service provider’s conversion capabilities. The conversion efficiency score is the ratio of a service provider’s conversion volume in the client’s industry to the service provider’s exposure in the client’s industry.

[0156] In some embodiments, the cost achievement rate score is the cost achievement advertising expenditure of all ads by the service provider in the client's industry plus the total expenditure of all ads by the service provider in the client's industry.

[0157] In some embodiments, the traffic expansion capability score is the service provider's spending on traffic expansion advertising in the customer's industry and the service provider's recharge amount for traffic expansion advertising in the customer's industry.

[0158] Since the service provider recommendation system is mainly used for new customers, the scaling-up cost and scaling-up speed are the most important indicators for new customers. If only the entropy weight method is used, the weight of the latter three indicators may be much higher than the scaling-up speed and scaling-up cost. Therefore, it is necessary to take customer preferences into account and ensure the rationality of the weights.

[0159] In some embodiments, the data consists of the names of shortlisted service providers and five columns of operational metrics. First, the data in each column is standardized using formula (8):

[0160] (8);

[0161] in, It is the standardized metric value of the j-th metric for the i-th recommended service object. It is the value of the j-th metric for the i-th recommended service object. It is the minimum value among multiple metrics for the i-th recommended service object. It is the maximum value among multiple metrics of the i-th recommended service object.

[0162] In some embodiments, the weight of the j-th indicator value of the i-th recommended service object to that indicator is calculated using formula (9):

[0163] (9);

[0164] in, It is the individual score of the j-th metric for the i-th recommended service object. It is the standardized index value of the j-th indicator for the i-th recommended service object.

[0165] In some embodiments, the information entropy of each indicator is calculated using formula (10):

[0166] (10);

[0167] in, It is the information entropy of the j-th index. It is the individual score of the j-th metric for the i-th recommended service object. It is the logarithmic result of the individual score of the i-th recommended service object corresponding to the j-th indicator. This is the second summation result, where n is the number of recommended service recipients.

[0168] In some embodiments, the information entropy value is optimized based on business experience, the weights of scaling speed and scaling loss are increased, and it is ensured that the weight of either scaling speed or scaling loss is the maximum value of the five indicators, see formula (11) and formula (12):

[0169] (11);

[0170] (12);

[0171] in, It is the optimized information entropy of the j-th index. It refers to the information entropy of the first indicator and the information entropy of the second indicator. It is the minimum value between the information entropy of the first indicator and the information entropy of the second indicator. These are the information entropy of the third, fourth, and fifth indicators. It is the minimum value among the information entropy of the third, fourth, and fifth indicators.

[0172] In some embodiments, the weights are calculated using formula (13):

[0173] (13);

[0174] in, It is the weight of the j-th indicator. It is the optimized information entropy of the j-th index, where k is a constant. It is the sum of the optimized information entropy of all indicators.

[0175] In some embodiments, the comprehensive score of the indicator is calculated using formula (14):

[0176] (14);

[0177] in, It is the overall score of the i-th recommended service recipient. It is the individual score of the j-th metric for the i-th recommended service object. It is the weight of the j-th indicator.

[0178] The final result is a comprehensive score of each service provider's operational metrics, and the service providers are ranked in descending order based on the comprehensive score.

[0179] In some embodiments, the ranking truncation module performs TOPM truncation on the descending score results and suggests that the client select N service providers for advertising. M is calculated as the average number of service providers used by clients in the target client's industry and level based on historical data. N is calculated as the number of service providers that cover 80% of the consumption among clients in the target client's industry and level based on historical data. This is because some service providers account for a large proportion of the client's consumption, while others only capture a small share. Therefore, it is recommended that the client select service providers that cover 80% of the historical data. Too many service providers would lead to excessive communication costs for the client.

[0180] At this point, the service provider recommendation list is generated, and the service provider's name, scores for each capability, overall rating, cooperation method, and other information are displayed to the client.

[0181] This application proposes a method for recommending service providers to customers with different needs. The method defines a suitable list of service providers based on the customer's industry, annual budget, traffic preferences, and support capabilities. Then, the target service providers are ranked either directly based on their financial capacity or based on their operational capabilities across various dimensions and their weights (related to customer preferences). Finally, the method recommends several top-ranked service providers to the customer based on their size and suggests that the customer select several of these service providers for advertising.

[0182] This application can comprehensively consider the personalized needs of the target users' indicator preferences and the objective indicator scores of the recommended service users. This can ensure that the recommended service users are well matched with the actual situation and needs of the target users, thereby improving the accuracy and efficiency of the recommendation.

[0183] It is understood that in the embodiments of this application, data such as user information are involved. When the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0184] The following description continues to illustrate the exemplary structure of the recommended processing apparatus 455 provided in the embodiments of this application as a software module. In some embodiments, such as Figure 2As shown, the software modules stored in the recommendation processing device 455 in the memory 450 may include: an acquisition module 4551, used to acquire first attribute information of the object to be served, wherein the first attribute information is used to characterize the degree of preference of the object to be served for each indicator of the recommended service object; an indicator module 4552, used to determine the individual scores of multiple indicators of each recommended service object; the indicator module 4552 is further used to determine the information entropy of each indicator based on the individual scores of multiple indicators of multiple recommended service objects, and to determine the weight of each indicator based on the information entropy of each indicator and the first attribute information; the indicator module 4552 is further used to perform weighted processing on the individual scores of multiple indicators of each recommended service object according to the weight of each indicator, to obtain the comprehensive score of the recommended service object; and a sorting module 4553, used to sort the multiple recommended service objects in a first descending order according to the comprehensive score of each recommended service object, and to perform a recommendation operation to the object to be served according to the first descending order sorting result.

[0185] In some embodiments, the acquisition module 4551 is further configured to: acquire second attribute information of the object to be served before determining the individual scores of multiple indicators for each recommended service object; perform at least one first screening process on multiple candidate recommended service objects based on the second attribute information to obtain a first screening result; acquire third attribute information of the object to be served; query candidate recommended service objects that match the third attribute information of the object to be served from the first screening result to form a target screening result set, wherein the third attribute information is used to characterize the type information of the object to be served; and determine at least one recommended service object that matches the object to be served based on the target screening result set.

[0186] In some embodiments, when the second attribute information includes industry information of the object to be served, the acquisition module 4551 is further configured to: acquire the first filtering object of the first filtering process, and perform the following processing for each candidate recommended service object in the first filtering object: determine the total revenue of the candidate recommended service object, the revenue of the industry information corresponding to the candidate recommended service object, and the revenue of the corresponding industry information; determine a first industry indicator that is negatively correlated with the total revenue of the candidate recommended service object and positively correlated with the revenue of the industry information corresponding to the candidate recommended service object; determine a second industry indicator that is negatively correlated with the revenue of the corresponding industry information and positively correlated with the revenue of the industry information corresponding to the candidate recommended service object; when the first industry indicator is greater than a first threshold and the second industry indicator is greater than a second threshold, retain the candidate recommended service object.

[0187] In some embodiments, when the second attribute information includes at least one level information corresponding to the object to be served, the acquisition module 4551 is further configured to: acquire the first filtering object of the first filtering process, and perform the following processing for each candidate recommended service object in the first filtering object: for each level information, determine the income of the industry information and level information corresponding to the candidate recommended service object, the income of the industry information and level information corresponding to the corresponding industry information, the income of the industry information corresponding to the corresponding industry information, and the income of the industry information corresponding to the candidate recommended service object; for each level information, determine the level index that is positively correlated with the income of the industry information and level information corresponding to the candidate recommended service object, positively correlated with the income of the industry information corresponding to the candidate recommended service object, negatively correlated with the income of the industry information and level information corresponding to the candidate recommended service object, and negatively correlated with the income of the industry information and level information corresponding to the candidate recommended service object; when the level index obtained for any level information satisfies the level preference condition, the candidate recommended service object is retained.

[0188] In some embodiments, the acquisition module 4551 is further configured to: determine the candidate recommended service objects in the target screening result set as at least one recommended service object matching the object to be served when the type information representation sorts the matched candidate recommended service objects using a non-independent index sorting method and the number of matched candidate recommended service objects does not exceed a number threshold; determine that the direct recommendation condition is met when the type information representation sorts the matched candidate recommended service objects using a non-independent index sorting method and the number of matched candidate recommended service objects does not exceed a number threshold; and perform a recommendation operation on the object to be served based on the candidate recommended service objects in the target screening result set when the direct recommendation condition is met.

[0189] In some embodiments, the acquisition module 4551 is further configured to: determine the candidate recommended service objects in the target screening result set as at least one recommended service object matching the object to be served when the type information representation will sort the matched candidate recommended service objects using an independent index sorting method; determine that the independent index sorting condition is met when the type information representation will sort the matched candidate recommended service objects using an independent index sorting method; when the independent index sorting condition is met, determine the individual scores of the corresponding type information indicators of multiple recommended service objects, sort the multiple recommended service objects in a second descending order based on the individual scores, and perform a recommendation operation on the object to be served based on the second descending order sorting result.

[0190] In some embodiments, when the type information representation will sort the matched candidate recommended service objects using a non-independent index sorting method, and the number of matched candidate recommended service objects exceeds the number threshold, the acquisition module 4551 is further configured to: acquire the fourth attribute information of the object to be served, and perform at least one second filtering process on the target filtering result set based on the fourth attribute information to obtain at least one recommended service object that matches the object to be served.

[0191] In some embodiments, the fourth attribute information includes at least one traffic channel information of the object to be served. The acquisition module 4551 is further configured to: acquire the second filtering object of the second filtering process, and perform the following processing for each candidate recommended service object in the second filtering object: for each traffic channel information, determine the revenue of the corresponding industry information and traffic channel information of the candidate recommended service object, the revenue of the corresponding industry information and traffic channel information, the revenue of the corresponding industry information, and the revenue of the corresponding industry information of the candidate recommended service object; for each traffic channel information, determine the traffic indicators that are positively correlated with the revenue of the corresponding industry information and traffic channel information of the candidate recommended service object, positively correlated with the revenue of the corresponding industry information, negatively correlated with the revenue of the corresponding industry information and traffic channel information of the candidate recommended service object, and negatively correlated with the revenue of the corresponding industry information and traffic channel information; when the traffic indicators obtained for each traffic channel information meet the channel preference conditions, the candidate recommended service object is retained.

[0192] In some embodiments, the acquisition module 4551 is further configured to: acquire the second filtering object of the second filtering process, and perform the following processing for each candidate recommended service object in the second filtering object: when the candidate recommended service object meets each capability requirement information, the candidate recommended service object is retained.

[0193] In some embodiments, the indicator module 4552 is further configured to: determine the indicator value of each indicator for each recommended service object; standardize the indicator value of each indicator for each recommended service object to obtain a standardized indicator value; for each indicator, sum the standardized indicator values ​​of multiple recommended service objects to obtain a first summation result, and determine the proportion of the standardized indicator value of each recommended service object to the first summation result as a separate score for each recommended service object's corresponding indicator.

[0194] In some embodiments, the indicator module 4552 is further configured to: perform the following processing for each indicator: perform logarithmic processing on the individual score of the indicator corresponding to each recommended service object, and multiply the logarithmic processing result with the individual score of the indicator corresponding to the recommended service object; sum the results of the multiplication of multiple recommended service objects to obtain a second summation result; and determine the information entropy of the indicator that is negatively correlated with the second summation result.

[0195] In some embodiments, the indicator module 4552 is further configured to: sort multiple indicators in descending order according to the degree of preference for each indicator; determine at least one indicator ranked first as a preferred indicator and the remaining indicators as non-preferred indicators; perform the following processing for each preferred indicator: determine an optimized information entropy that is positively correlated with the information entropy of the preferred indicator and a first minimum information entropy, and negatively correlated with a second minimum information entropy; wherein the first minimum information entropy is the minimum information entropy among the information entropies of multiple indicators, and the second minimum information entropy is the minimum information entropy among the information entropies of at least one preferred indicator; perform the following processing for each non-preferred indicator: determine the information entropy of the non-preferred indicator as the optimized information entropy of the corresponding non-preferred indicator; sum the optimized information entropies of multiple indicators to obtain a total information entropy; and perform the following processing for each indicator: determine the weight of the indicator that is negatively correlated with the optimized information entropy of the indicator and positively correlated with the total information entropy.

[0196] In some embodiments, the sorting module 4553 is further configured to: obtain industry information and grade information of the object to be served; query multiple historical objects to be served corresponding to the industry information and grade information; determine the number of historical recommended service objects for each historical object to be served; average the number of historical recommended service objects for multiple historical objects to be served to obtain an average number; obtain the recommended service objects that are ranked first in the first descending sorting result and meet the average number; and perform a recommendation operation on the object to be served based on the recommended service objects that meet the average number.

[0197] This application provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the recommended processing method described above in this application.

[0198] This application provides a computer-readable storage medium storing executable instructions. When these executable instructions are executed by a processor, they cause the processor to perform the recommended processing method provided in this application, for example... Figures 3A-3C The recommended processing method is shown.

[0199] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or it may be a variety of devices including one or any combination of the above-mentioned memories.

[0200] In some embodiments, executable instructions may take the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a standalone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0201] As an example, executable instructions may, but do not necessarily, correspond to files in a file system. They may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a Hyper Text Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple collaborating files (e.g., a file that stores one or more modules, subroutines, or code sections).

[0202] As an example, executable instructions can be deployed to execute on a single computing device, or on multiple computing devices located in one location, or on multiple computing devices distributed across multiple locations and interconnected via a communication network.

[0203] In summary, the embodiments of this application can comprehensively consider the personalized needs of the target user's indicator preferences and the objective indicator scores of the recommended service users. This ensures that the recommended service users are well matched with the target user's actual situation and needs, thereby improving the accuracy and efficiency of the recommendation.

[0204] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, and improvements made within the spirit and scope of this application are included within the scope of protection of this application.

Claims

1. A recommended processing method, characterized in that, The method includes: Obtain the first attribute information of the object to be served, wherein the first attribute information is used to characterize the degree of preference of the object to be served for each indicator of the recommended service object; Determine individual scores for each of the aforementioned metrics for each of the recommended service objects; Based on the individual scores of multiple metrics for multiple recommended service objects, the information entropy of each metric is determined, and the weight of each metric is determined based on the information entropy of each metric and the first attribute information. For each of the recommended service objects, the individual scores of multiple indicators of the recommended service object are weighted according to the weight of each indicator to obtain the comprehensive score of the recommended service object; Based on the comprehensive score of each of the recommended service objects, the multiple recommended service objects are sorted in a first descending order, and a recommendation operation is performed on the object to be served based on the first descending order result.

2. The method according to claim 1, characterized in that, Before determining individual scores for the multiple metrics for each of the recommended service objects, the method further includes: Obtain the second attribute information of the object to be served, and perform at least one first filtering process on multiple candidate recommended service objects based on the second attribute information to obtain the first filtering result; Obtain the third attribute information of the object to be served, query the first filtering result for candidate recommended service objects that match the third attribute information of the object to be served, and form a target filtering result set, wherein the third attribute information is used to characterize the type information of the object to be served; Based on the target filtering result set, at least one recommended service object that matches the object to be served is determined.

3. The method according to claim 2, characterized in that, When the second attribute information includes the industry information of the object to be served, any one of the first filtering processes includes: Obtain the first filtering object from the first filtering process, and perform the following processing for each candidate recommendation service object in the first filtering object: Determine the total revenue of the candidate recommendation service object, the revenue of the candidate recommendation service object corresponding to the industry information, and the revenue corresponding to the industry information; A first industry indicator is determined that is negatively correlated with the total revenue of the candidate recommended service object and positively correlated with the revenue of the industry information corresponding to the candidate recommended service object; A second industry indicator is determined that is negatively correlated with the revenue corresponding to the industry information and positively correlated with the revenue corresponding to the industry information of the candidate recommendation service object; When the first industry indicator is greater than the first threshold and the second industry indicator is greater than the second threshold, the candidate recommended service object is retained.

4. The method according to claim 2, characterized in that, When the second attribute information includes the industry information of the object to be served and at least one level information corresponding to the object to be served, any one of the first filtering processes includes: Obtain the first filtering object from the first filtering process, and perform the following processing for each candidate recommendation service object in the first filtering object: For each level of information, determine the revenue of the candidate recommended service object corresponding to the industry information and the level information, the revenue corresponding to the industry information and the level information, the revenue corresponding to the industry information, and the revenue of the candidate recommended service object corresponding to the industry information; For each level information, determine the level index that is positively correlated with the income of the industry information and the level information corresponding to the candidate recommended service object, positively correlated with the income of the industry information, negatively correlated with the income of the industry information corresponding to the candidate recommended service object, and negatively correlated with the income of the industry information and the level information. When the level index obtained for any of the level information meets the level preference condition, the candidate recommended service object is retained.

5. The method according to claim 2, characterized in that, The method further includes: When the type information representation uses a non-independent index ranking method to sort the matched candidate recommended service objects, and the number of matched candidate recommended service objects does not exceed a threshold, determining at least one recommended service object matching the object to be served based on the target screening result set includes: The candidate recommended service objects in the target screening result set are determined as at least one recommended service object that matches the object to be served; The method further includes: When the type information representation sorts the matched candidate recommendation service objects using the non-independent index sorting method, and the number of matched candidate recommendation service objects does not exceed the number threshold, it is determined that the direct recommendation condition is met. When the direct recommendation conditions are met, a recommendation operation is performed on the object to be served based on the candidate recommendation service objects in the target screening result set.

6. The method according to claim 2, characterized in that, When the type information representation uses an independent index ranking method to rank the matched candidate recommended service objects, the step of determining at least one recommended service object matching the object to be served based on the target screening result set includes: The candidate recommendation service objects in the target screening result set are determined as multiple recommendation service objects that match the object to be served; The method further includes: When the type information representation sorts the matched candidate recommendation service objects using the independent index sorting method, it is determined that the independent index sorting condition is met. When the independent index ranking condition is met, a separate score for the index corresponding to the type information of multiple recommended service objects is determined, and the multiple recommended service objects are sorted in a second descending order based on the separate scores. A recommendation operation is then performed on the service object based on the second descending order ranking result.

7. The method according to claim 2, characterized in that, When the type information representation uses a non-independent index ranking method to rank the matched candidate recommended service objects, and the number of matched candidate recommended service objects exceeds a threshold, the step of determining at least one recommended service object matching the object to be served based on the target screening result set includes: Obtain the fourth attribute information of the object to be served, and perform at least one second filtering process on the target filtering result set based on the fourth attribute information to obtain at least one recommended service object that matches the object to be served.

8. The method according to claim 7, characterized in that, When the second attribute information includes the industry information of the object to be served, and the fourth attribute information includes at least one traffic channel information of the object to be served, any one of the second filtering processes includes: Obtain the second filtering object from the second filtering process, and perform the following processing for each candidate recommendation service object in the second filtering object: For each of the traffic channel information, determine the revenue of the candidate recommendation service object corresponding to the industry information and the traffic channel information, the revenue corresponding to the industry information and the traffic channel information, the revenue corresponding to the industry information, and the revenue of the candidate recommendation service object corresponding to the industry information; For each of the traffic channel information, determine the traffic metrics that are positively correlated with the revenue of the industry information and the traffic channel information corresponding to the candidate recommendation service object, positively correlated with the revenue of the industry information, negatively correlated with the revenue of the industry information corresponding to the candidate recommendation service object, and negatively correlated with the revenue of the industry information and the traffic channel information. When the traffic metrics obtained for each of the traffic channels meet the channel preference conditions, the candidate recommended service objects are retained.

9. The method according to claim 7, characterized in that, The fourth attribute information includes at least one capability requirement information of the object to be served, and any one of the second filtering processes includes: Obtain the second filtering object from the second filtering process, and perform the following processing for each candidate recommendation service object in the second filtering object: When a candidate recommended service object meets each of the capability requirements, the candidate recommended service object is retained.

10. The method according to claim 1, characterized in that, The process of determining individual scores for each of the multiple metrics for each of the recommended service objects includes: Determine the index value for each of the aforementioned metrics for each of the recommended service objects; The index values ​​of each of the aforementioned indicators for each of the recommended service objects are standardized to obtain standardized index values; For each of the aforementioned indicators, the standardized indicator values ​​of multiple recommended service objects corresponding to the indicator are summed to obtain a first summation result. The proportion of the standardized indicator value of each recommended service object corresponding to the indicator in the first summation result is determined as a separate score for each recommended service object corresponding to the indicator.

11. The method according to claim 1, characterized in that, The step of determining the information entropy of each of the multiple metrics based on individual scores of the multiple recommended service objects includes: For each of the aforementioned metrics, perform the following processing: Logarithmic processing is performed on the individual score of each recommended service object corresponding to the indicator, and the logarithmic processing result is multiplied by the individual score of the corresponding recommended service object corresponding to the indicator. The results of multiplying multiple recommended service objects are summed to obtain a second summation result; Determine the information entropy of the index that is negatively correlated with the second summation result.

12. The method according to claim 1, characterized in that, The step of determining the weight of each indicator based on the information entropy of each indicator and the first attribute information includes: Based on the degree of preference for each of the indicators, the multiple indicators are sorted in descending order; At least one of the top-ranked indicators is identified as a preferred indicator, and the remaining indicators are identified as non-preferred indicators. For each of the preference indicators, the following processing is performed: Determine the optimized information entropy that is positively correlated with the information entropy of the preference indicator and the first minimum information entropy, and negatively correlated with the second minimum information entropy; Wherein, the first minimum information entropy is the minimum information entropy among the information entropies of the plurality of said indicators, and the second minimum information entropy is the minimum information entropy among the information entropies of at least one said preference indicator; For each of the non-preference indicators, the following process is performed: the information entropy of the non-preference indicator is determined as the optimized information entropy of the corresponding non-preference indicator; The optimized information entropy of multiple indicators is summed to obtain the total information entropy; For each of the metrics, the following process is performed: determine the weight of the metric that is negatively correlated with the optimized information entropy of the metric and positively correlated with the sum of the information entropies.

13. The method according to claim 1, characterized in that, The step of performing a recommendation operation on the objects to be served based on the first descending sort result includes: Obtain the industry information and rating information of the target service recipient; Query multiple historical pending service objects corresponding to the industry information and the grade information; Determine the number of historical recommended service objects for each of the historical pending service objects; The average number of historical recommended service objects for multiple historical pending service objects is calculated by averaging the number of such objects. Obtain the recommended service objects that meet the average number of sorted items at the top of the first descending sort result; Based on the number of recommended service recipients that meet the average number, a recommendation operation is performed on the recipients to be served.

14. A recommended processing apparatus, characterized in that, The device includes: The acquisition module is used to acquire the first attribute information of the object to be served, wherein the first attribute information is used to characterize the degree of preference of the object to be served for each indicator of the recommended service object; The metrics module is used to determine individual scores for multiple metrics for each of the recommended service objects; The indicator module is further configured to determine the information entropy of each indicator based on the individual scores of multiple indicators of multiple recommended service objects, and to determine the weight of each indicator based on the information entropy of each indicator and the first attribute information. The indicator module is also used to perform weighted processing on the individual scores of multiple indicators of each recommended service object according to the weight of each indicator, so as to obtain the comprehensive score of the recommended service object. The sorting module is used to sort the multiple recommended service objects in a first descending order based on the comprehensive score of each recommended service object, and to perform a recommendation operation on the service object according to the first descending order result.

15. The apparatus according to claim 14, characterized in that, The acquisition module is further configured to acquire the second attribute information of the object to be served before determining the individual scores of the multiple indicators of each of the recommended service objects, and perform at least one first screening process on the multiple candidate recommended service objects based on the second attribute information to obtain the first screening result; Obtain the third attribute information of the object to be served, query the first filtering result for candidate recommended service objects that match the third attribute information of the object to be served, and form a target filtering result set, wherein the third attribute information is used to characterize the type information of the object to be served; Based on the target filtering result set, at least one recommended service object that matches the object to be served is determined.

16. The apparatus according to claim 15, characterized in that, When the second attribute information includes the industry information of the object to be served. The acquisition module is further configured to acquire the first filtering object of the first filtering process, and perform the following processing for each candidate recommendation service object in the first filtering object: Determine the total revenue of the candidate recommendation service object, the revenue of the candidate recommendation service object corresponding to the industry information, and the revenue corresponding to the industry information; A first industry indicator is determined that is negatively correlated with the total revenue of the candidate recommended service object and positively correlated with the revenue of the industry information corresponding to the candidate recommended service object; A second industry indicator is determined that is negatively correlated with the revenue corresponding to the industry information and positively correlated with the revenue corresponding to the industry information of the candidate recommendation service object; When the first industry indicator is greater than the first threshold and the second industry indicator is greater than the second threshold, the candidate recommended service object is retained.

17. The apparatus according to claim 15, characterized in that, When the second attribute information includes the industry information of the object to be served and at least one level information corresponding to the object to be served. The acquisition module is further configured to acquire the first filtering object of the first filtering process, and perform the following processing for each candidate recommendation service object in the first filtering object: For each level of information, determine the revenue of the candidate recommended service object corresponding to the industry information and the level information, the revenue corresponding to the industry information and the level information, the revenue corresponding to the industry information, and the revenue of the candidate recommended service object corresponding to the industry information; For each level information, determine the level index that is positively correlated with the income of the industry information and the level information corresponding to the candidate recommended service object, positively correlated with the income of the industry information, negatively correlated with the income of the industry information corresponding to the candidate recommended service object, and negatively correlated with the income of the industry information and the level information. When the level index obtained for any of the level information meets the level preference condition, the candidate recommended service object is retained.

18. The apparatus according to claim 15, characterized in that, The acquisition module is further configured to determine the candidate recommended service objects in the target screening result set as at least one recommended service object that matches the object to be served when the type information representation will sort the matched candidate recommended service objects by a non-independent index sorting method and the number of matched candidate recommended service objects does not exceed the number threshold. The acquisition module is further configured to determine that the direct recommendation condition is met when the type information representation will sort the matched candidate recommendation service objects using the non-independent index sorting method, and the number of matched candidate recommendation service objects does not exceed the number threshold. When the direct recommendation conditions are met, a recommendation operation is performed on the object to be served based on the candidate recommendation service objects in the target screening result set.

19. The apparatus according to claim 15, characterized in that, The acquisition module is further configured to determine the candidate recommended service objects in the target screening result set as multiple recommended service objects that match the object to be served when the type information representation will sort the matched candidate recommended service objects by an independent index sorting method; The acquisition module is further configured to determine that the independent index ranking condition is met when the type information representation will rank the matched candidate recommendation service objects using the independent index ranking method. When the independent index ranking condition is met, the individual scores of the index corresponding to the type information of multiple recommended service objects are determined, the multiple recommended service objects are sorted in a second descending order according to the individual scores, and the recommendation operation is performed on the service object according to the second descending order ranking result.

20. The apparatus according to claim 15, characterized in that, When the type information representation uses a non-independent index ranking method to rank the matched candidate recommendation service objects, and the number of matched candidate recommendation service objects exceeds a threshold, The acquisition module is further configured to acquire the fourth attribute information of the object to be served, and perform at least one second filtering process on the target filtering result set based on the fourth attribute information to obtain at least one recommended service object that matches the object to be served.

21. The apparatus according to claim 20, characterized in that, When the second attribute information includes the industry information of the object to be served, and the fourth attribute information includes at least one traffic channel information of the object to be served. The acquisition module is further configured to acquire the second filtering object of the second filtering process, and perform the following processing for each candidate recommendation service object in the second filtering object: For each of the traffic channel information, determine the revenue of the candidate recommendation service object corresponding to the industry information and the traffic channel information, the revenue corresponding to the industry information and the traffic channel information, the revenue corresponding to the industry information, and the revenue of the candidate recommendation service object corresponding to the industry information; For each of the traffic channel information, determine the traffic metrics that are positively correlated with the revenue of the industry information and the traffic channel information corresponding to the candidate recommendation service object, positively correlated with the revenue of the industry information, negatively correlated with the revenue of the industry information corresponding to the candidate recommendation service object, and negatively correlated with the revenue of the industry information and the traffic channel information. When the traffic metrics obtained for each of the traffic channels meet the channel preference conditions, the candidate recommended service objects are retained.

22. The apparatus according to claim 20, characterized in that, The fourth attribute information includes at least one capability requirement of the object to be served. The acquisition module is further configured to acquire the second filtering object of the second filtering process, and perform the following processing for each candidate recommendation service object in the second filtering object: When a candidate recommended service object meets each of the capability requirements, the candidate recommended service object is retained.

23. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; A processor, when executing executable instructions stored in the memory, implements the recommended processing method according to any one of claims 1 to 13.

24. A computer-readable storage medium storing executable instructions, characterized in that, When the executable instructions are executed by the processor, they implement the recommended processing method according to any one of claims 1 to 13.

25. A computer program product comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the recommended processing method according to any one of claims 1 to 13.

Images