Data processing method and device, computer equipment and computer readable storage medium

Abstract

The invention discloses a data processing method and device, computer equipment and a computer readable storage medium, the method is applied to scenes such as cloud technology and artificial intelligence, and the method comprises the following steps: obtaining fan data, work data and price data of a first account; determining index values of N evaluation indexes by a first sub-model in the target evaluation model according to fan data, work data and price data of the first account; determining an index value of a comprehensive index by a second sub-model in the target evaluation model according to the index values of the N evaluation indexes; and the index values of the N evaluation indexes and the index value of the comprehensive index are stored in association with the first account. In the embodiment of the invention, not only the basic data (namely fan data, work data and price data) of the account exists, but also the index data determined by the basic data of the account exists, so that the richness of the data can be improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and more specifically, to a data processing method, apparatus, computer equipment, and computer-readable storage medium. Background Technology

[0002] With the continuous development of computer technology, various content publishing platforms have emerged. Many content publishing platforms store various data about the accounts on the platform so that they can filter accounts to meet various needs based on this data. However, currently, the data used for filtering accounts stored by content publishing platforms includes basic data such as the number of followers and pricing information, which is relatively limited. Summary of the Invention

[0003] In view of the above problems, embodiments of this application propose a data processing method, apparatus, computer equipment, and computer-readable storage medium to improve the above problems.

[0004] In a first aspect, embodiments of this application disclose a data processing method, including:

[0005] Obtain the first account's follower data, work data, and price data;

[0006] The first sub-model in the target evaluation model determines the values ​​of N evaluation indicators based on the fan data, the work data, and the price data; N is an integer greater than 1.

[0007] The second sub-model in the target evaluation model determines the value of the comprehensive indicator based on the values ​​of the N evaluation indicators.

[0008] The values ​​of the N evaluation indicators and the value of the comprehensive indicator are associated with and stored with the first account.

[0009] Secondly, embodiments of this application disclose a data processing apparatus, including:

[0010] The acquisition unit is used to acquire the first account's fan data, work data, and price data;

[0011] The first determining unit is used to determine the index values ​​of N evaluation indicators by the first sub-model in the target evaluation model based on the fan data, the work data, and the price data; N is an integer greater than 1.

[0012] The first determining unit is further configured to determine the index value of the comprehensive index by the second sub-model in the target evaluation model based on the index values ​​of the N evaluation indicators;

[0013] The storage unit is used to associate and store the index values ​​of the N evaluation indicators and the index value of the comprehensive indicator with the first account.

[0014] Thirdly, embodiments of this application disclose a computer device, including: a processor; a memory, wherein computer-readable instructions are stored in the memory, and when the computer-readable instructions are executed by the processor, the method described above is implemented.

[0015] Fourthly, embodiments of this application disclose a computer-readable storage medium storing computer-readable instructions thereon, which, when executed by a processor, implement the method described above.

[0016] Sixthly, embodiments of this application disclose a computer program product, including computer instructions, which, when executed by a processor, implement the method described above.

[0017] In this embodiment, the follower data, work data, and price data of a first account are obtained. A first sub-model in the target evaluation model determines the values ​​of N evaluation indicators based on these data. A second sub-model in the target evaluation model determines the value of a comprehensive indicator based on the values ​​of the N evaluation indicators. The values ​​of the N evaluation indicators and the comprehensive indicator are then associated and stored with the first account. It is evident that the values ​​of the N evaluation indicators and the comprehensive indicator for an account can be determined based on the account's basic data (i.e., follower data, work data, and price data). Therefore, it not only provides the account's basic data but also the indicator data determined by this basic data. The N evaluation indicators correspond to different evaluation dimensions, and the comprehensive indicator value is the account's overall evaluation value determined by referring to the indicator values ​​under the N evaluation indicators. This provides not only indicator values ​​under different evaluation dimensions but also a comprehensive evaluation value that reflects the account's overall situation, thereby improving the richness of the data used for account screening. Furthermore, since the account's indicator data is determined by the target evaluation model based on the account's basic data, the efficiency of indicator data determination can be improved. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0019] Figure 1 This is a schematic diagram of a network architecture disclosed in an embodiment of this application;

[0020] Figure 2This is a flowchart illustrating a data processing method disclosed in an embodiment of this application;

[0021] Figure 3 This is a flowchart illustrating another data processing method disclosed in an embodiment of this application;

[0022] Figure 4 This is a flowchart illustrating another data processing method disclosed in the embodiments of this application;

[0023] Figure 5 This is a flowchart illustrating another data processing method disclosed in the embodiments of this application;

[0024] Figure 6 This is a schematic diagram of a target evaluation model structure disclosed in an embodiment of this application;

[0025] Figure 7 This is a schematic diagram of an AHP model constructed according to an embodiment of this application;

[0026] Figure 8 This is a schematic diagram of an application scenario disclosed in an embodiment of this application;

[0027] Figure 9 This is a schematic diagram of a client-side interface disclosed in an embodiment of this application;

[0028] Figure 10 This is a schematic diagram illustrating another application scenario disclosed in the embodiments of this application;

[0029] Figure 11 This is a schematic diagram showing the indicator values ​​and rankings of various indicators of an expert, as disclosed in an embodiment of this application.

[0030] Figure 12 This is a schematic diagram of the structure of a data processing apparatus disclosed in an embodiment of this application;

[0031] Figure 13 This is a schematic diagram of the structure of a computer system of a computer device disclosed in an embodiment of this application. Detailed Implementation

[0032] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make this application more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art.

[0033] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this application. However, those skilled in the art will recognize that the technical solutions of this application can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this application.

[0034] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware models or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0035] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0036] It should be noted that "multiple" in this article refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0037] To better understand the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained below.

[0038] The Analytic Hierarchy Process (APH) treats a complex multi-objective decision problem as a system. It decomposes the objective into multiple objectives or criteria, and further into several levels of multiple indicators (or criteria, constraints). Using qualitative index fuzzy quantification methods, it calculates the hierarchical single ranking (weights) and overall ranking, serving as a systematic method for optimizing decisions across multiple objectives and alternatives. Specifically, it decomposes the decision problem into different hierarchical structures according to the overall objective, sub-objectives at each level, evaluation criteria, and finally, specific alternative solutions. Then, it uses the method of solving the eigenvectors of the judgment matrix to obtain the weight of each element at each level with respect to an element at the previous level. Finally, it uses a weighted sum method to hierarchically merge the final weights of each alternative solution with respect to the overall objective; the solution with the largest final weight is the optimal solution.

[0039] Normalization: Normalization is a data processing method that restricts data to a fixed range after processing. In other words, normalization can convert dimensional data into dimensionless data and normalize data to the same order of magnitude, solving the problem of data comparability.

[0040] Coefficient of Variation (CV) method: The CV method is an objective weighting method that calculates the degree of change of each indicator in a system based on statistical methods. It assigns weights to each indicator based on the degree of variation between its current value and the target value. When the current value of an indicator differs significantly from the target value, it indicates that the indicator is unlikely to achieve the target value and should be given a larger weight; conversely, it should be given a smaller weight.

[0041] Influencer: In this application, an influencer refers to a high-quality account that is determined by a comprehensive evaluation of multiple factors, including the quality of the works published on the content publishing application, the popularity of the works, the dissemination of the works, and the update frequency of the works.

[0042] To better understand the embodiments of this application, the relevant technologies of this application will be introduced below.

[0043] Different platforms have accumulated a large amount of data on various accounts. This data can be used to evaluate accounts effectively and efficiently, allowing for the quick and accurate selection of accounts that meet specific needs.

[0044] However, the platform provides basic data such as the number of followers, likes, comments, and pricing information for accounts. This data is relatively limited, making it impossible to quickly and accurately evaluate and filter based on this basic data.

[0045] To address the aforementioned issues, this application discloses a data processing method that can determine the values ​​of N evaluation indicators and comprehensive indicators for an account based on its basic data (i.e., fan data, work data, and price data). Therefore, it can provide not only the basic data of the account but also the indicator data determined by the basic data, thereby improving the richness of the account data.

[0046] To better understand the embodiments of this application, the network architecture of this application is described below.

[0047] Please see Figure 1 , Figure 1 This is a schematic diagram of a network architecture disclosed in an embodiment of this application. For example... Figure 1As shown, the network architecture may include a computer device 101, multiple terminal devices 102, and at least one content publishing application server 103. The computer device 101 can be connected to the multiple terminal devices 102 via a network. The computer device 101 can also be connected to at least one content publishing application server 103 via a network. The network can be a wide area network (WAN), a local area network (LAN), or a combination of both. The content publishing application server 103 refers to a server that provides services for content publishing applications. Figure 1 The example shows two content publishing application servers 103, which can correspond to different content publishing applications.

[0048] A client is installed on terminal device 102. This client is the client corresponding to computer device 101. The client can communicate with computer device 101 through terminal device 102. When a user needs to connect to computer device 101, the user can operate terminal device 102 to launch the client. After detecting the user's launch operation to launch the client, terminal device 102 can respond to the launch operation and display the client's login interface. After detecting the user's login operation to log in to computer device 101 through the login interface, terminal device 102 can respond to the login operation and send account information to computer device 101. After receiving the account information, computer device 101 can verify whether the account corresponding to the account information is a legitimate account. If the account corresponding to the account information is verified to be a legitimate account, it can send data to terminal device 102.

[0049] After receiving data from computer device 101, terminal device 102 can display a data interface. The data interface may include data about the account. This data may include at least one of the following: follower data, work data, and price data.

[0050] The client in terminal device 102 can, based on user operations, communicate with computer device 101 to modify works of the account, adjust the price of works of the account, publish new works, and delete works of the account.

[0051] The data interface can also include data from other accounts. The client in terminal device 102 can view, browse, watch, play, like, etc., data from other accounts by communicating with computer device 101 based on user operations.

[0052] The accounts logged in through the aforementioned client are those registered on computer device 101. Computer device 101 can store all data for these accounts.

[0053] The computer device 101 can also periodically retrieve and store the basic data of the corresponding account from the content publishing application server 103. The account corresponding to the content publishing application server 103 is the account registered on the content publishing application server 103. The basic data of the account here can be the latest basic data of the account. For example, the computer device 101 can retrieve the basic data of the corresponding account for the current day from the content publishing application server 103 every day.

[0054] Computer device 101 can periodically or periodically determine the indicator values ​​of each indicator of the account based on the account's basic data, and can store the indicator values ​​of each indicator of the account.

[0055] In some instances, computer device 101 can send the metric values ​​of various account metrics to terminal device 102. After receiving the metric values ​​of various account metrics from computer device 101, terminal device 102 can display the metric values ​​of various account metrics through a client.

[0056] In some embodiments, the computer device 101 can sort accounts according to the index values ​​of each metric to obtain the ranking of the accounts under each metric, and then store the ranking of the accounts under each metric. The accounts here can be accounts from the same channel, accounts from different regions of the same channel, or all accounts that the computer device 101 can obtain.

[0057] In some embodiments, computer device 101 may send the account's ranking under various metrics to terminal device 102. After receiving the account's ranking under various metrics from computer device 101, terminal device 102 can display the account's ranking under various metrics through a client.

[0058] Computer device 101 may or may not be a content publishing application server.

[0059] In some embodiments, the computer device can be a terminal device or a server. The terminal device or server can implement the data 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; it can also be a small program, that is, a program that only needs to be downloaded into a browser environment to run; or it can be a small program that can be embedded in any APP, wherein the small program can be controlled by the user to run or close. In summary, the above-mentioned computer program can be any form of application, module or plugin.

[0060] In some embodiments, the content publishing application server 103 can be an independent 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, content delivery networks (CDNs), and big data and artificial intelligence platforms. Terminal devices can be smartphones, tablets, laptops, desktop computers, smart TVs, smartwatches, in-vehicle terminals, etc., but are not limited to these. Terminal devices and servers can be directly or indirectly connected via wired or wireless communication, which is not limited in this embodiment.

[0061] It should be understood that Figure 1 This is an exemplary illustration of the network architecture of this application and does not limit the network architecture of this application.

[0062] The implementation details of the technical solutions in the embodiments of this application are described in detail below.

[0063] Based on the above network architecture, please refer to Figure 2 , Figure 2 This is a schematic flowchart illustrating a data processing method disclosed in an embodiment of this application. This data processing method can be applied to the aforementioned computer equipment. Figure 2 As shown, the data processing method may include the following steps.

[0064] 201. Obtain the fan data, work data, and price data of the first account.

[0065] The first account refers to any account already registered in a content publishing application. A content publishing application is an application with content publishing capabilities. For example, it could be a video publishing application, an article publishing application, a note publishing application, an image publishing application, an audio publishing application, etc. The solution in this application can process data related to a registered account in one content publishing application or to registered accounts in multiple content publishing applications (i.e., the follower data, work data, and price data mentioned above).

[0066] The system can periodically or periodically acquire the basic data of the first account, so that the indicator values ​​of each indicator of the first account can be determined based on the basic data of the first account. Then, the indicator values ​​of each indicator of the first account in the computer device can be periodically or periodically updated, so that the latest indicator values ​​of each indicator of the first account are stored, which can ensure the real-time and validity of the indicator values.

[0067] In the specific implementation of this application, data such as account follower data, work data and price data are involved. When the various embodiments of this application are applied to specific products or technologies, the collection, use and processing of relevant data (such as account follower data, work data and price data) need to comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0068] The basic data for the first account can include the account's follower data, work data, and price data.

[0069] The follower data for Account 1 refers to the follower-related data of Account 1. This data can include the number of followers for Account 1 at different times. It can also include the number of followers for Account 1 at two different times. The number of followers for Account 1 refers to the number of accounts following Account 1, i.e., the number of accounts following Account 1. For example, if the desired follower growth rate is the follower growth rate over the past 30 days, the follower data for Account 1 can be as shown in Table 1:

[0070] Time (days) The value of the number of fans Day 1 A1 Day 30 A2

[0071] Table 1 Fan Data

[0072] If the follower data for the first account today exists, then today is the 30th day. If the follower data for the first account today does not exist, then yesterday is the 30th day. A1 and A2 are non-negative integers.

[0073] The data for the first account's works refers to data related to the works of the first account. This data can include information about the works published by the first account. Published work data can include one or more of the following: number of views (or page views), publication time, update time, number of likes, number of comments, number of favorites, number of shares, and playback duration (or viewing duration). The number of views is the total number of times the work has been viewed by other accounts. Works can be videos, articles, images, etc. If the work is an article, opening the article indicates that the work has been viewed. If the work is an image, opening the image indicates that the work has been played. The publication time is the time the work was uploaded or published. The update time is the time the work was modified after publication. The number of likes is the total number of likes the work has received. For example, the data for the first account's works can include data from the first account's latest 15 works, as shown in Table 2.

[0074] artwork logo Release time Update time The value of play count The value of likes Work 1 Time 1 / B1 C1 Work 2 Time 2 Time 16 / C2 Work 3 Time 3 / B2 / ﹒﹒﹒ ﹒﹒﹒ ﹒﹒﹒ ﹒﹒﹒ ﹒﹒﹒ Work 15 Time 15 / B /

[0075] Table 2. Works Data

[0076] In this context, the " / " below the update time indicates no update time, meaning the work has not been updated since its publication. If the work has been updated multiple times, the update time can include multiple entries. The " / " below the view count value indicates no views (0 views). The " / " below the like count value indicates no likes (0 likes).

[0077] The pricing data for the first account refers to the pricing-related data for that account. This data can include the quoted price and the cost per thousand impressions (CPM). The quoted price is the price offered by the first account for undertaking tasks (such as orders, advertisements, etc.). The quoted price can be the latest value or all possible values. The quoted price can be a single value or a range. The CPM value is the expected CPM value for the first account. The pricing data for the first account can be shown in Table 3.

[0078] logo value Quoted Price D1 CPM D2

[0079] Table 3 Price Data

[0080] 202. The first sub-model in the target evaluation model determines the values ​​of N evaluation indicators based on the fan data, work data and price data of the first account.

[0081] 203. The second sub-model in the target evaluation model determines the value of the comprehensive indicator based on the values ​​of the N evaluation indicators.

[0082] A target evaluation model is a pre-trained model used to determine the values ​​of various metrics for an account based on its follower data, content data, and price data. A target evaluation model can include a first sub-model and a second sub-model.

[0083] The first sub-model in the target evaluation model can determine the values ​​of N evaluation indicators based on the first account's follower data, work data, and price data. That is, the first sub-model can input the first account's follower data, work data, and price data, and the output of the first sub-model is the value of the first account's N evaluation indicators. N is an integer greater than 1.

[0084] Then, the second sub-model in the target evaluation model can determine the value of the comprehensive indicator based on the values ​​of the N evaluation indicators. That is, the values ​​of the N evaluation indicators of the first account output by the first sub-model can be input into the second sub-model, and the output of the second sub-model is the value of the comprehensive indicator of the first account.

[0085] The N evaluation metrics are evaluation indicators set under N evaluation dimensions for assessing an account, with one evaluation dimension corresponding to one evaluation indicator. Evaluation dimensions can be influence evaluation dimensions, work update evaluation dimensions, work virality evaluation dimensions, work dissemination evaluation dimensions, account cost-effectiveness evaluation dimensions, etc. The indicator value for each evaluation indicator is determined by at least one of the following: fan data, work data, and price data.

[0086] The N evaluation metrics can include fan metrics, update metrics, hit product metrics, dissemination metrics, and cost-effectiveness metrics.

[0087] The follower metric is an indicator used to assess an account's influence within the influence evaluation dimension, reflecting the account's overall impact. The follower metric is determined based on the account's follower data. The follower metric value is positively correlated with the number of followers; the more followers an account has, the higher its follower metric value. The follower metric value is also positively correlated with the account's follower growth rate.

[0088] Account update metrics are indicators used to evaluate the update activity of an account's works within the work update evaluation dimension, reflecting the account's work update status. The value of the account update metric is determined based on the account's work data. Specifically, the account update metric can be determined based on the view count and release time of multiple works by the account, or it can be determined based on the view count, release time, and update time of multiple works by the account.

[0089] The "Viral Content Metric" for an account is used to evaluate whether a post from an account has become a viral hit, within the broader category of viral content evaluation. The metric value can be determined based on the view count of multiple posts from the account, or on the combined view count and like count of multiple posts. A higher Viral Content Metric indicates a higher viral rate for the account's posts.

[0090] The account's dissemination metric is an indicator used to evaluate the dissemination power of the works published by an account, reflecting the overall dissemination status of the accounts' published works. The metric value is determined based on the account's follower data and work data. Specifically, the metric value can be determined based on the account's follower data and the views of multiple works, or it can be determined based on the account's follower data and the views and likes of multiple works.

[0091] The account cost-effectiveness metric is used to evaluate the cost-effectiveness of an account within the overall account cost-effectiveness assessment dimension. It reflects the cost-effectiveness of using the account to undertake tasks. The value of the account cost-effectiveness metric is determined based on the account's pricing data.

[0092] The comprehensive index of an account is determined based on the index values ​​of N evaluation indicators of the account. It is used to reflect the overall situation of the account, that is, to reflect the comprehensive capabilities of the account.

[0093] The above index values ​​can be values ​​located in [0,1], [0,100], [0,1000], or other intervals.

[0094] 204. The values ​​of the N evaluation indicators and the value of the comprehensive indicator are associated with and stored with the first account.

[0095] After determining the values ​​of the N evaluation indicators and the comprehensive indicator for the first account, these values ​​can be associated with and stored for later retrieval. The values ​​of the N indicators and the comprehensive indicator for the first account are shown in Table 4.

[0096]

[0097] Table 4 Indicator Values

[0098] exist Figure 2 The described data processing method determines the values ​​of N evaluation indicators and a comprehensive indicator for the account based on its basic data (i.e., follower data, content data, and price data). This demonstrates that it not only provides the account's basic data but also the indicator data determined by that data. The N evaluation indicators correspond to different evaluation dimensions, and the comprehensive indicator value is a holistic evaluation value for the account determined by referencing the values ​​of the N evaluation indicators. This provides not only indicator values ​​across different evaluation dimensions but also a comprehensive evaluation value that reflects the account's overall performance, thus enriching the account's data. Furthermore, since the account's indicator data is determined by the target evaluation model based on the account's basic data, the efficiency of indicator data determination is improved.

[0099] Based on the above network architecture, see Figure 3 , Figure 3 This is a schematic flowchart of another data processing method disclosed in an embodiment of this application. This data processing method can be applied to the aforementioned computer equipment. Figure 3 As shown, the data processing method may include the following steps.

[0100] 301. Obtain training data including fan data, work data, price data, and tag data from multiple accounts.

[0101] If you need to train an evaluation model, you can first obtain training data. Training data can include follower data, work data, price data, and tag data from multiple accounts.

[0102] Multiple accounts can be registered accounts for at least one content publishing application.

[0103] For a detailed description of the follower data, work data, and price data for multiple accounts, please refer to the description of the follower data, work data, and price data for the first account below step 201. The follower data, work data, and price data for different accounts may be the same or different.

[0104] The tag data for these multiple accounts includes the tag value of N evaluation metrics and the tag value of the comprehensive metric for these accounts. The tag values ​​of the N evaluation metrics for these multiple accounts are pre-determined values ​​for each of the N evaluation metrics for these multiple accounts. The tag value of the comprehensive metric for these multiple accounts is also pre-determined. Both the tag values ​​of the N evaluation metrics and the tag value of the comprehensive metric for these multiple accounts are determined based on the conversion performance of these multiple accounts.

[0105] For a detailed description of the N evaluation indicators and comprehensive indicators, please refer to the relevant description below step 203.

[0106] It should be understood that in order to ensure the accuracy of the trained evaluation model, the training data must include a sufficiently large number of samples, that is, the number of accounts from multiple accounts must be large enough.

[0107] 302. The first initial sub-model in the initial evaluation model determines the index values ​​of N evaluation indicators for the second account based on the fan data, work data, and price data of the second account.

[0108] 303. The second initial sub-model in the initial evaluation model determines the comprehensive index value of the second account based on the index values ​​of the N evaluation indicators of the second account.

[0109] The second account can be any one of the aforementioned accounts. The initial evaluation model may include a first initial sub-model and a second initial sub-model.

[0110] For a detailed description of steps 302-303, please refer to the description of steps 202-203.

[0111] 304. Determine the loss value based on the label and indicator values ​​of the N evaluation indicators of the second account, as well as the label and indicator values ​​of the comprehensive indicator.

[0112] After determining the values ​​of the N evaluation indicators and the comprehensive indicator for the second account, the loss value can be determined based on the tag values ​​and indicator values ​​of the N evaluation indicators and the tag values ​​and indicator values ​​of the comprehensive indicator for the second account.

[0113] Based on the tag values ​​and indicator values ​​of the N indicators of the second account, the loss value corresponding to each of the N evaluation indicators can be determined. In other words, based on the tag values ​​and indicator values ​​of the N indicators of the second account, the loss value corresponding to each of the N evaluation indicators can be determined, resulting in N loss values. There is a one-to-one correspondence between the N evaluation indicators and the N loss values.

[0114] The overall loss value can be determined based on the tag value and indicator value of the comprehensive indicator of the second account. That is, the loss value corresponding to the comprehensive indicator is determined based on the tag value and indicator value of the comprehensive indicator of the second account, and the overall loss value is obtained.

[0115] There are no restrictions on the type of loss function used to calculate the loss value. For example, the loss function can be the cross-entropy loss function, the L1 (absolute error) loss function, the L2 (mean squared error) loss function, the smoothing loss function, etc.

[0116] 305. Optimize the parameters of the initial evaluation model based on the loss value to obtain the target evaluation model.

[0117] Once the loss value is determined, the parameters of the initial evaluation model can be optimized based on the loss value to obtain the target evaluation model. For example, the parameters of the initial evaluation model can be optimized using the back propagation (BP) algorithm.

[0118] The weight parameters of the first initial sub-model can be optimized based on the loss values ​​corresponding to N evaluation indicators, and the weight parameters of the second initial sub-model can be optimized based on the comprehensive loss value.

[0119] The above method first determines the values ​​of N evaluation indicators and the comprehensive indicator for the account. Then, based on the label indicator values ​​and the comprehensive indicator values ​​of the account's N evaluation indicators, the loss value is determined. The parameters of the initial evaluation model are updated based on the loss value, which enables effective updating of the parameters of the initial evaluation model and ensures the accuracy of subsequent indicator value prediction using the target evaluation model.

[0120] 306. Obtain the fan data, work data, and price data of the first account.

[0121] For a detailed description of step 306, please refer to the description of step 201.

[0122] 307. The first sub-model in the target evaluation model determines the values ​​of N evaluation indicators based on the fan data, work data and price data of the first account.

[0123] 308. The second sub-model in the target evaluation model determines the value of the comprehensive indicator based on the values ​​of N evaluation indicators.

[0124] For a detailed description of steps 307-308, please refer to the description of steps 202-203.

[0125] 309. Associate and store the values ​​of the N evaluation indicators and the comprehensive indicator with the first account.

[0126] For a detailed description of step 309, please refer to the description of step 204.

[0127] exist Figure 3 The described data processing method first trains a target evaluation model using training data. Then, it uses this trained model to determine the values ​​of N evaluation metrics and a comprehensive metric for the account. This eliminates the need for retraining the evaluation model every time it's used, improving the efficiency of determining the account's metric values. Based on the account's basic data (i.e., follower data, content data, and price data), the method determines the values ​​of the account's N evaluation metrics and comprehensive metric. Therefore, it not only provides the account's basic data but also the metric values ​​of the N evaluation metrics corresponding to different evaluation dimensions determined by this basic data, as well as the comprehensive metric value determined based on the values ​​of the N evaluation metrics, thus enriching the account's data.

[0128] Based on the above network architecture, see Figure 4 , Figure 4 This is a schematic flowchart illustrating another data processing method disclosed in an embodiment of this application. This data processing method can be applied to the aforementioned computer equipment. Figure 4 As shown, the data processing method may include the following steps.

[0129] 401. Obtain training data including fan data, work data, price data, and tag data from multiple accounts.

[0130] For a detailed description of step 401, please refer to the description of step 301.

[0131] 402. Obtain fan data, work data, and price data for K accounts.

[0132] Before training the evaluation model, it is necessary to determine the initial parameters of the initial evaluation model. Therefore, to determine the initial parameters of the initial evaluation model, we can obtain the follower data, work data, and price data of K accounts. K is an integer greater than 1.

[0133] K accounts can be all accounts from one or more channels.

[0134] The K accounts can be any K accounts from the multiple accounts mentioned above. In this way, to train the evaluation model, no additional data besides the training data is needed, reducing the amount of data acquired and thus saving storage and processing resources. Accordingly, steps 402-408 are executed after step 401.

[0135] The K accounts may not be among the accounts mentioned above; that is, the data used to determine the initial parameters of the initial evaluation model may differ from the training data used to train the initial evaluation model. Accordingly, steps 402-408 and step 401 can be executed sequentially or in parallel.

[0136] For a detailed description of the follower data, work data, and price data of the K accounts, please refer to the relevant description of the follower data, work data, and price data of the first account below step 201.

[0137] 403. Based on the fan data, work data, and price data of each of the K accounts, determine the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts.

[0138] After obtaining the follower data, work data, and price data of K accounts, the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts can be determined based on these data. In other words, the parameter value of each of the M basic parameters corresponding to each of the K accounts can be determined based on the follower data, work data, and price data of each account. M is an integer greater than N.

[0139] First, based on the fan data, work data, and price data of each of the K accounts, determine the values ​​of each of the M basic parameters corresponding to each of the K accounts.

[0140] Since the values ​​of the M basic parameters may have different dimensions, in order to eliminate the influence of the different dimensions, the values ​​of the M basic parameters can be processed so that the processed values ​​of the M basic parameters are dimensionless. Therefore, the values ​​of each of the M basic parameters corresponding to each of the K accounts can be normalized to obtain the parameter value of each of the M basic parameters corresponding to each of the K accounts.

[0141] In some embodiments, after obtaining the follower data, work data, and price data of K accounts, the data can be cleaned first. Then, based on the cleaned follower data, work data, and price data of each of the K accounts, the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts can be determined. Data cleaning may include data improvement and data filtering.

[0142] Data improvement involves processing and correcting data to ensure its accuracy and completeness. For example, if the data for a third account's work includes the number of likes but not the number of views, data improvement can be used to determine the number of likes as the number of views. The third account can be any one of K accounts.

[0143] Data filtering involves selecting data that meets specific requirements, i.e., removing data that does not. For example, you could remove the number of likes on a third-party account's work. Or, you could select data from multiple works by a third-party account.

[0144] For example, the M basic parameters may include the number of followers, follower growth rate, video update rate, video views, video virality rate, video stability, CPM, and pricing. The following provides a detailed explanation of the parameter values ​​for determining the number of followers, follower growth rate, video update rate, video views, video virality rate, and video stability.

[0145] The latest value of the number of followers in the follower data of the third account can be determined as the value of the number of followers of the third account. In other words, the value of the number of followers in the follower data of the third account corresponding to the most recent time can be determined as the value of the number of followers of the third account.

[0146] The follower growth rate of a third account can be determined based on its follower data. This growth rate is calculated as the increase in followers over a preset time period, which is the ratio of the increase in followers to a reference value. The increase in followers is the difference between the number of followers at the first time point and the number of followers at the second time point. The first time point is the end time of the preset time period, and the second time point is the start time of the preset time period. The reference value for the number of followers is the number of followers at the second time point.

[0147] For example, the fan growth rate value within a preset time period can be the fan growth rate value of the most recent 30 days. Accordingly, the account's fan growth rate value = (fan count value on day 30 - fan count value on day 1) / fan count value on day 1.

[0148] The update rate of a third account can be determined based on its work data. Specifically, it can be determined by the publication and update times of multiple works from the third account. The average update time interval of multiple works from the third account can also be used to determine the update rate. The average update time interval of multiple works from an account is the average update time interval of adjacent works. The update time interval between two works can be the time interval between their publication times, the time interval between their update times, or the time interval between the publication time of one work and the update time of another.

[0149] For example, the update rate of an account can be the average of the update intervals of the account's most recent 15 works. First, the account's most recent 15 works can be sorted from earliest to latest to obtain a work sorting table. Then, the update intervals between adjacent works in the work sorting table can be calculated to obtain 14 time intervals. The average of these 14 time intervals can be calculated to obtain the account's work update rate.

[0150] The view count of a third account's works can be determined based on its work data. Specifically, the view count of a third account's works can be determined based on the view counts of multiple works by that account. The view count of an account's works can be the average view count of multiple works by that account, the median view count of multiple works by that account, or the total view count of multiple works by that account.

[0151] For example, the view count of an account's works can be the median of the view counts of the account's most recent 15 works. Assuming that the view counts of the account's most recent 15 works are F1, F2, ..., F15, we can sort F1, F2, ..., F15 in ascending or descending order to obtain a view count sorting table. The value at the 8th position in the view count sorting table (i.e., the median of the view counts of the most recent 15 works) can be determined as the view count of the account's works.

[0152] The viral rate of a third account's works can be determined based on its work data. Specifically, the viral rate can be determined by the view count of multiple works from the third account. The viral rate can be calculated as the ratio of the number of works whose view counts rank within the top 90 percentile of the view counts of K accounts to the total number of works. The K accounts can include all works from the K accounts, or multiple works from each of the K accounts. For example, the viral rate of an account = the number of works from that account whose view counts rank within the top 90 percentile of the view counts of the K accounts / 15. The threshold percentile can be understood as a percentage threshold. Works whose view counts rank within the top 90 percentile of the view counts of the K accounts can be understood as works whose view counts rank within the top 90 percentile of the view counts of the K accounts. For example, if the K accounts have 100 works and the threshold percentile is 90 percentile, works whose view counts rank within the top 90 percentile of the view counts of the K accounts can be understood as works whose view counts rank within the top 90 percentile of the view counts of the K accounts.

[0153] In other embodiments, works that meet the criteria for becoming a viral hit can be selected from the works of a third account. These are the works considered viral hits from the third account's output. The ratio of the number of viral hits from the third account to the total number of works from the third account is then determined as the viral hit rate of that third account. The criteria for becoming a viral hit could be a first preset threshold for views and a second preset threshold for shares. Of course, the criteria could also be limited to the number of comments, the number of favorites, etc., but these are not specifically limited here.

[0154] The stability value of a third-party account's content can be determined based on its content data. Specifically, the stability value can be determined based on the play counts of multiple works by the third-party account. The coefficient of variation of the play counts of multiple works by the third-party account can be used to determine the stability value of the third-party account's content. For example, the coefficient of variation of the play counts of the third-party account's most recent 15 works can be used to determine the stability value of the third-party account's content.

[0155] The average view count of multiple works from a third-party account can be determined, and the standard deviation of the view count can also be determined. Based on the average and standard deviation of view counts, the stability value of the third-party account's works can be determined. Specifically, the ratio of the standard deviation to the average view count can be used to determine the stability value of the third-party account's works.

[0156] The CPM and quoted price of a third-party account can be determined based on its price data. If the third-party account's price data includes multiple CPM values, the latest CPM value from that data can be used to determine the account's CPM value. If the third-party account's price data includes only one CPM value, that value can be used to determine the account's CPM value. If the quoted price in the third-party account's price data is a single value, that value can be used to determine the account's quoted price value. If the quoted price in the third-party account's price data is a range, the midpoint of that range can be used to determine the account's quoted price value. Therefore, the CPM and quoted price values ​​of a third-party account are determined based on its latest price data.

[0157] Since the values ​​of number of fans, fan growth rate, video update rate, video views, video hit rate, video stability, CPM, and quoted price have different dimensions, in order to eliminate the impact of the different dimensions, the values ​​of number of fans, fan growth rate, video update rate, video views, video hit rate, video stability, CPM, and quoted price can be normalized to obtain the parameter values ​​of number of fans, fan growth rate, video update rate, video views, video hit rate, video stability, CPM, and quoted price.

[0158] The parameter value for the number of followers of a third account can be represented as follows:

[0159]

[0160] Where K1 represents the parameter value of the number of followers of the third account, and x1 represents the value of the number of followers of the third account. max x represents the maximum value among the number of followers of K accounts. 1min This represents the minimum number of followers among the K accounts.

[0161] The parameter value for the fan growth rate of a third account can be expressed as follows:

[0162]

[0163] Where K2 represents the parameter value of the third account's fan growth rate, and x2 represents the value of the third account's fan growth rate. max This represents the maximum value among the fan growth rates of K accounts, x2 minThis represents the minimum value among the fan growth rates of K accounts.

[0164] The parameter value for the update rate of a third account's works can be represented as follows:

[0165]

[0166] Where K3 represents the parameter value of the third account's update rate, and x3 represents the value of the third account's update rate. max This represents the maximum value among the update rates of K accounts, x3 min This represents the minimum value among the update rates of works from K accounts.

[0167] The view count parameter for a third account's work can be represented as follows:

[0168]

[0169] Where K4 represents the view count of the third account's work, and x4 represents the view count of the third account's work. max This represents the maximum value among the view counts of works from K accounts, x4 min This represents the minimum number of views among the works from K accounts.

[0170] The parameter value for the popularity rate of works from a third account can be expressed as follows:

[0171]

[0172] Where K5 represents the parameter value of the third account's work's hit rate, and x5 represents the value of the third account's work's hit rate. max This represents the maximum value among the K accounts' works' hit rate, x5 min This represents the minimum value among the K accounts' works' hit rate.

[0173] The parameter values ​​for the stability of works created by a third account can be represented as follows:

[0174]

[0175] Where K6 represents the parameter value of the stability of the third account's works, and x6 represents the value of the stability of the third account's works. max This represents the maximum value among the stability values ​​of the works from K accounts, x6 min This represents the minimum value among the stability values ​​of works from K accounts.

[0176] The CPM parameter value for a third account can be represented as follows:

[0177]

[0178] Where K7 represents the CPM parameter value of the third account, and x7 represents the CPM value of the third account. max This represents the maximum CPM value among K accounts, x7 min This represents the minimum CPM value among the K accounts.

[0179] The parameter values ​​for the quoted price of a third account can be represented as follows:

[0180]

[0181] Where K8 represents the parameter value of the price quoted by the third account, and x8 represents the value of the price quoted by the third account. max This represents the maximum value among the bid prices of K accounts, x8 min This represents the minimum value among the quoted prices from K accounts.

[0182] 404. Based on the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts, determine the initial weights corresponding to the M basic parameters respectively.

[0183] After determining the parameter values ​​of each of the M basic parameters for each of the K accounts, the initial weights corresponding to the M basic parameters can be determined based on the parameter values ​​of each of the M basic parameters for each of the K accounts.

[0184] The initial weights corresponding to the M basic parameters include the initial weights of each of the N evaluation indicators corresponding to each of the M basic parameters. We can determine the basic parameters related to the target evaluation indicator from the M basic parameters, thus obtaining multiple basic parameters corresponding to the target evaluation indicator. We can determine the average value of the target basic parameter for each of the K accounts, thus obtaining the average value of the target basic parameter corresponding to the target evaluation indicator. We can also determine the standard deviation of the parameter values ​​for each of the K accounts, thus obtaining the standard deviation of the target basic parameter corresponding to the target evaluation indicator. Based on the average value and standard deviation of the target basic parameter corresponding to the target evaluation indicator, we determine the initial weights of the target basic parameter corresponding to the target evaluation indicator, thus obtaining the initial weights of each of the N evaluation indicators corresponding to each of the M basic parameters. The target evaluation indicator is any one of the N evaluation indicators. The target basic parameter is any one of the multiple basic parameters corresponding to the target evaluation indicator.

[0185] The following section provides a detailed explanation of the initial weights for each of the following metrics: number of fans, fan growth rate, update rate of works, number of views of works, rate of works becoming popular, and stability of works, corresponding to the fan metric, update metric, popular metric, dissemination metric, and cost-effectiveness metric.

[0186] This method allows you to determine the basic parameters related to fan metrics, including fan count, fan growth rate, video update rate, video views, viral video rate, video stability, CPM, and pricing, to obtain the corresponding fan count and fan growth rate. It also allows you to determine the basic parameters related to update metrics, including fan count, fan growth rate, video update rate, video views, viral video rate, video stability, CPM, and pricing, to obtain the corresponding video update rate and video views. Finally, it allows you to determine the basic parameters related to dissemination metrics, including fan count, fan growth rate, video update rate, video views, viral video rate, video stability, CPM, and pricing, to obtain the corresponding fan count, fan growth rate, video views, and video stability. It can determine the basic parameters related to cost-effectiveness indicators among the number of fans, fan growth rate, work update rate, work views, work hit rate, work stability, CPM and quoted price, and obtain the corresponding CPM and quoted price for cost-effectiveness indicators.

[0187] We can determine the average value of the number of followers for K accounts, and we can also determine the standard deviation of the number of followers for K accounts. The ratio of the average value to the standard deviation can be defined as the coefficient of variation of the number of followers. The coefficient of variation of the number of followers can be expressed as follows:

[0188]

[0189] in, y represents the average value of the number of followers of K accounts. j Let S represent the parameter value of the number of followers for the j-th account, S represent the standard deviation of the parameter values ​​of the number of followers for the K accounts, and V represent the coefficient of variation of the number of followers.

[0190] The average value of the parameter value of the fan growth rate of K accounts can be determined, and the standard deviation of the parameter value of the fan growth rate of K accounts can be determined. The ratio of the average value to the standard deviation is determined as the coefficient of variation of the fan growth rate.

[0191] The average value of the parameter value of the update rate of K accounts can be determined, and the standard deviation of the parameter value of the update rate of K accounts can be determined. The ratio of the average value to the standard deviation is used to determine the coefficient of variation of the update rate of the work.

[0192] The average value of the view count parameter for K accounts can be determined, as can the standard deviation of the view count parameter for K accounts. The ratio of the average value to the standard deviation is used to determine the coefficient of variation of the view count.

[0193] The average value of the hit rate parameter for K accounts can be determined, as can the standard deviation of the hit rate parameter for K accounts. The ratio of the average value to the standard deviation can be used to determine the coefficient of variation of the hit rate.

[0194] The average value of the parameter values ​​for the stability of works from K accounts can be determined, as can the standard deviation of the parameter values ​​for the stability of works from K accounts. The ratio of the average value to the standard deviation can be used to determine the coefficient of variation of the stability of the works.

[0195] The average value of the CPM parameter for K accounts can be determined, as can the standard deviation of the CPM parameter value for K accounts. The ratio of the average value to the standard deviation can be determined as the coefficient of variation of CPM.

[0196] The average value of the parameter values ​​of the quoted prices of K accounts can be determined, and the standard deviation of the parameter values ​​of the quoted prices of K accounts can be determined. The ratio of the average value to the standard deviation can be determined as the coefficient of variation of the quoted prices.

[0197] The coefficients of variation for the number of followers and the coefficient of variation for the follower growth rate can be normalized to obtain the initial weights of the follower metrics corresponding to the number of followers and the follower growth rate. The initial weight of the follower metric corresponding to the number of followers can be determined by the ratio of the coefficient of variation for the number of followers to the sum of the coefficients of variation for the number of followers and the coefficient of variation for the follower growth rate.

[0198] The coefficients of variation of the update rate and the number of views can be normalized to obtain the initial weights of the update metrics corresponding to the update rate and the number of views. The initial weight of the update metric corresponding to the update rate can be determined by the ratio of the coefficient of variation of the update rate to the sum of the coefficients of variation of the update rate and the number of views.

[0199] The coefficients of variation of a work's viral hit rate and its viewership can be normalized to obtain the initial weights of these metrics. The initial weight of the viral hit rate metric can be determined by the ratio of its coefficient of variation to the sum of the coefficients of variation of its viral hit rate and its viewership. Similarly, the initial weight of the viewership metric can be determined by the ratio of its coefficient of variation to the sum of the coefficients of variation of its viral hit rate and its viewership.

[0200] The coefficients of variation for follower count, follower growth rate, video views, and video stability can be normalized to obtain the initial weights for these corresponding dissemination metrics. The initial weight for the follower count dissemination metric can be determined by the ratio of the coefficient of variation for follower count to the sum of these coefficients. Similarly, the initial weight for the follower growth rate dissemination metric can be determined by the ratio of the coefficient of variation for follower count to the sum of these coefficients. The initial weight for the video views dissemination metric can be determined by the ratio of the coefficient of variation for video stability to the sum of these coefficients.

[0201] The coefficients of variation of CPM and quoted price can be normalized to obtain the initial weights of the cost-effectiveness indicators corresponding to CPM and quoted price. The initial weight of the cost-effectiveness indicator corresponding to CPM can be determined by the ratio of the coefficient of variation of CPM to the sum of the coefficients of variation of quoted price and vice versa.

[0202] 405. Based on the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts, and the initial weights corresponding to the M basic parameters, determine the indicator values ​​of each of the N evaluation indicators corresponding to each of the K accounts.

[0203] Based on the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts, and the initial weights corresponding to the M basic parameters, the indicator values ​​of each of the N evaluation indicators corresponding to the K accounts can be determined.

[0204] The following section provides a detailed explanation of the values ​​for each of the following metrics: fan metrics, update metrics, hit product metrics, dissemination metrics, and cost-effectiveness metrics.

[0205] The sum of the product of the number of followers of the third account and the initial weight of the follower metric corresponding to the number of followers, and the product of the parameter value of the follower growth rate of the third account and the initial weight of the follower metric corresponding to the follower growth rate, can be used to determine the metric value of the follower metric corresponding to the third account.

[0206] The sum of the product of the update rate parameter of the third account and the initial weight of the update indicator corresponding to the update rate, and the product of the view count parameter of the third account and the initial weight of the update indicator corresponding to the view count, can be used to determine the indicator value of the update indicator corresponding to the third account.

[0207] The product of the third account's hit rate parameter value and the initial weight of the hit rate corresponding to the hit indicator can be used to determine the indicator value of the third account's corresponding hit indicator.

[0208] The index value of the third account's corresponding dissemination index can be determined by multiplying the parameter value of the number of followers of the third account by the initial weight of the dissemination index corresponding to the number of followers, the parameter value of the third account's follower growth rate by the initial weight of the dissemination index corresponding to the follower growth rate, the parameter value of the third account's work stability by the initial weight of the dissemination index corresponding to the work stability, and the parameter value of the third account's work views by the initial weight of the dissemination index corresponding to the work views.

[0209] The value of the cost-effectiveness indicator for the third account can be determined by multiplying the CPM parameter value of the third account by the initial weight of the cost-effectiveness indicator corresponding to the CPM, and by multiplying the parameter value of the third account's quoted price by the initial weight of the cost-effectiveness indicator corresponding to the quoted price.

[0210] 406. Based on the index values ​​of each of the N evaluation indicators corresponding to the K accounts, determine the initial weights corresponding to the N evaluation indicators.

[0211] The initial weights of the N evaluation indicators can be determined based on the indicator values ​​of the N evaluation indicators corresponding to the K accounts.

[0212] We can determine the average value of the target evaluation indicator for each of the K accounts, thus obtaining the average value of the target evaluation indicator. We can also determine the standard deviation of the target evaluation indicator for each of the K accounts, thus obtaining the standard deviation of the target evaluation indicator. Based on the average value and standard deviation of the target evaluation indicator, we can determine the initial weight of the target evaluation indicator, thus obtaining the initial weight of the N evaluation indicators.

[0213] The following section provides a detailed explanation of the initial weights for determining the fan metrics, update metrics, hit product metrics, dissemination metrics, and cost-effectiveness metrics.

[0214] The average value of the fan metrics for K accounts can be determined, as can the standard deviation of the fan metrics for K accounts. The ratio of the average value to the standard deviation can be defined as the coefficient of variation of the fan metrics.

[0215] The average value of the update metric for K accounts can be determined, as can the standard deviation of the update metric for K accounts. The ratio of the average value to the standard deviation can be defined as the coefficient of variation of the update metric.

[0216] The average value of the hit product indicator for K accounts can be determined, as can the standard deviation of the hit product indicator value for K accounts. The ratio of the average value to the standard deviation can be determined as the coefficient of variation of the hit product indicator.

[0217] The average value of the propagation metric for K accounts can be determined, as can the standard deviation of the propagation metric for K accounts. The ratio of the average value to the standard deviation can be defined as the coefficient of variation of the propagation metric.

[0218] The average value of the cost-effectiveness index for K accounts can be determined, as can the standard deviation of the cost-effectiveness index for K accounts. The ratio of the average value to the standard deviation can be determined as the coefficient of variation of the cost-effectiveness index.

[0219] The coefficients of variation of the fan metric, update metric, best-selling metric, dissemination metric, and cost-effectiveness metric can be normalized to obtain the initial weights corresponding to the fan metric, update metric, best-selling metric, dissemination metric, and cost-effectiveness metric, respectively.

[0220] The initial weight corresponding to the fan metric can be determined by the ratio of the coefficient of variation of the fan metric to the sum of the coefficients of variation of the update metric, the coefficient of variation of the hit product metric, the coefficient of variation of the dissemination metric, and the coefficient of variation of the cost-effectiveness metric.

[0221] The initial weight corresponding to the update indicator can be determined by the ratio of the coefficient of variation of the update indicator to the sum of the coefficients of variation of the fan indicator, the update indicator, the hit product indicator, the dissemination indicator, and the cost-effectiveness indicator.

[0222] The initial weight corresponding to the hit product metric can be determined by the ratio of the coefficient of variation of the hit product metric to the sum of the coefficients of variation of the fan metric, the update metric, the hit product metric, the dissemination metric, and the cost-effectiveness metric.

[0223] The initial weight corresponding to the dissemination metric can be determined by the ratio of the coefficient of variation of the dissemination metric to the sum of the coefficients of variation of the fan metric, the update metric, the hit product metric, the dissemination metric, and the cost-effectiveness metric.

[0224] The initial weight corresponding to the cost-effectiveness indicator can be determined by the ratio of the coefficient of variation of the cost-effectiveness indicator to the sum of the coefficients of variation of the fan indicator, the update indicator, the hit product indicator, the dissemination indicator, and the cost-effectiveness indicator.

[0225] For example, the indicator values ​​for 5 evaluation indicators corresponding to 3 accounts can be shown in Table 5:

[0226]

[0227]

[0228] Table 5 Indicator Values

[0229] Based on the data in Table 5, the initial weights of the following metrics were calculated using the above method: fan metrics 0.1, best-selling metrics 0.1, update metrics 0.1, dissemination metrics 0.3, and cost-effectiveness metrics 0.4.

[0230] It should be understood that methods such as the coefficient of variation method, empirical judgment method, principal component analysis method, and analytic hierarchy process can be used to determine the initial weights corresponding to the M basic parameters and the initial weights corresponding to the N evaluation indicators.

[0231] Steps 402-406 use the coefficient of variation method to determine the initial weights corresponding to the M basic parameters in the first initial sub-model and the initial weights corresponding to the N evaluation indicators in the second initial sub-model.

[0232] 407. Initialize the weights corresponding to the M basic parameters in the first initial sub-model to their respective initial weights, and initialize the weights corresponding to the N evaluation indicators in the second initial sub-model to their respective initial weights.

[0233] After determining the initial weights corresponding to the M basic parameters, the weights corresponding to the M basic parameters in the first initial sub-model can be initialized to their respective initial weights.

[0234] After determining the initial weights corresponding to the N evaluation indicators, the weights corresponding to the N evaluation indicators in the second initial sub-model can be initialized to their respective initial weights.

[0235] 408. The first initial sub-model in the initial evaluation model determines the index values ​​of N evaluation indicators for the second account based on the fan data, work data, and price data of the second account.

[0236] 409. The second initial sub-model in the initial evaluation model determines the comprehensive index value of the second account based on the index values ​​of the N evaluation indicators of the second account.

[0237] For a detailed description of steps 408-409, please refer to the description of steps 302-303.

[0238] 410. Determine the loss value based on the label and indicator values ​​of the N evaluation indicators of the second account, as well as the label and indicator values ​​of the comprehensive indicator.

[0239] For a detailed description of step 410, please refer to the description of step 304.

[0240] 411. Optimize the parameters of the initial evaluation model based on the loss value to obtain the target evaluation model.

[0241] For a detailed description of step 411, please refer to the description of step 305.

[0242] 412. Obtain the fan data, work data, and price data of the first account.

[0243] For a detailed description of step 412, please refer to the description of step 201.

[0244] 413. The first sub-model in the target evaluation model determines the values ​​of N evaluation indicators based on the fan data, work data and price data of the first account.

[0245] For a detailed description of step 413, please refer to the description of step 202.

[0246] 414. The second sub-model in the target evaluation model determines the value of the comprehensive indicator based on the values ​​of N evaluation indicators.

[0247] For a detailed description of step 414, please refer to the description of step 203.

[0248] 415. Associate and store the values ​​of the N evaluation indicators and the comprehensive indicator with the first account.

[0249] For a detailed description of step 415, please refer to the description of step 204.

[0250] exist Figure 4 The described data processing method first determines the initial parameters of the initial evaluation model, then trains the target evaluation model based on the training data, and uses the trained target evaluation model to determine the indicator values ​​of each indicator. This eliminates the need to determine parameters and train the evaluation model every time it is used, thus improving the efficiency of determining indicator values. Based on the account's basic data (i.e., follower data, work data, and price data), the indicator values ​​of N evaluation indicators and a comprehensive indicator for the account can be determined. Therefore, it not only provides the account's basic data but also the indicator values ​​of N evaluation indicators corresponding to different evaluation dimensions determined by the account's basic data, as well as the indicator values ​​of the comprehensive indicator determined based on the indicator values ​​of the N evaluation indicators, enriching the account's data.

[0251] Based on the above network architecture, see Figure 5 , Figure 5 This is a schematic flowchart of another data processing method disclosed in an embodiment of this application. This data processing method can be applied to the aforementioned computer equipment. Figure 5 As shown, the data processing method may include the following steps.

[0252] 501. Obtain training data including fan data, work data, price data, and tag data from multiple accounts.

[0253] For a detailed description of step 501, please refer to the description of step 301.

[0254] 502. The third sub-model in the initial evaluation model determines the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts based on the fan data, work data and price data of each account.

[0255] The K accounts are the K accounts among the aforementioned multiple accounts.

[0256] The third sub-model in the initial evaluation model can determine the parameter values ​​of each of the M basic parameters for each of the K accounts based on the follower data, work data, and price data of each account. Specifically, the follower data, work data, and price data of each of the K accounts are input into the third sub-model of the initial evaluation model, and the output of the third sub-model is the parameter value of each of the M basic parameters for each of the K accounts.

[0257] For a detailed description of step 502, please refer to the description of step 403.

[0258] 503. The fourth sub-model in the initial evaluation model determines the initial weights corresponding to the M basic parameters and the initial weights corresponding to the N evaluation indicators based on the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts.

[0259] The initial evaluation model's fourth sub-model determines the initial weights for the M basic parameters and the initial weights for the N evaluation indicators based on the parameter values ​​of the M basic parameters corresponding to each of the K accounts. Specifically, the parameter values ​​of the M basic parameters corresponding to each of the K accounts are input into the fourth sub-model of the initial evaluation model, and the output of the fourth sub-model is the initial weights for the M basic parameters and the initial weights for the N evaluation indicators.

[0260] For a detailed description of step 503, please refer to the descriptions of steps 404-406.

[0261] 504. Initialize the weights corresponding to the M basic parameters in the first initial sub-model to their respective initial weights, and initialize the weights corresponding to the N evaluation indicators in the second initial sub-model to their respective initial weights.

[0262] For a detailed description of step 504, please refer to the description of step 407.

[0263] 505. Based on the fan data, work data and price data of the second account, the third sub-model in the initial evaluation model determines the parameter values ​​of each of the M basic parameters corresponding to the second account.

[0264] For a detailed description of step 505, please refer to the description of step 502.

[0265] 506. The first initial sub-model in the initial evaluation model determines the index values ​​of N evaluation indicators for the second account based on the parameter values ​​of each of the M basic parameters corresponding to the second account.

[0266] The first initial sub-model in the initial evaluation model can determine the index values ​​of the N evaluation indicators of the second account based on the parameter values ​​of each of the M basic parameters corresponding to the second account. That is, the parameter values ​​of each of the M basic parameters corresponding to the second account are input into the first initial sub-model, and the output of the first initial sub-model is the index values ​​of the N evaluation indicators of the second account.

[0267] 507. The second initial sub-model in the initial evaluation model determines the comprehensive index value of the second account based on the index values ​​of the N evaluation indicators of the second account.

[0268] For a detailed description of step 507, please refer to the description of step 303.

[0269] 508. Determine the loss value based on the label and indicator values ​​of the N evaluation indicators of the second account, as well as the label and indicator values ​​of the comprehensive indicator.

[0270] For a detailed description of step 508, please refer to the description of step 304.

[0271] 509. Optimize the parameters of the initial evaluation model based on the loss value to obtain the target evaluation model.

[0272] For a detailed description of step 509, please refer to the description of step 305.

[0273] 510. Obtain the fan data, work data, and price data of the first account.

[0274] For a detailed description of step 510, please refer to the description of step 201.

[0275] 511. The third sub-model in the target evaluation model determines the parameter values ​​of each of the M basic parameters corresponding to the first account based on the fan data, work data and price data of the first account.

[0276] For a detailed description of step 513, please refer to the description of step 505.

[0277] 512. The first sub-model in the target evaluation model determines the index values ​​of N evaluation indicators based on the parameter values ​​of each of the M basic parameters corresponding to the first account.

[0278] For a detailed description of step 512, please refer to the description of step 506.

[0279] 513. The second sub-model in the target evaluation model determines the value of the comprehensive indicator based on the values ​​of N evaluation indicators.

[0280] For a detailed description of step 513, please refer to the description of step 203.

[0281] 514. Associate and store the values ​​of the N evaluation indicators and the comprehensive indicator with the first account.

[0282] For a detailed description of step 514, please refer to the description of step 204.

[0283] For example, see Figure 6 , Figure 6 This is a schematic diagram of a target evaluation model structure disclosed in an embodiment of this application. For example... Figure 6 As shown, the target evaluation model includes a first sub-model, a second sub-model, a third sub-model, and a fourth sub-model. The account's follower data, work data, and price data are input into the third sub-model. The third sub-model outputs the parameter values ​​of each of the M basic parameters corresponding to the account. The parameter values ​​of each of the M basic parameters corresponding to the account are input into the first sub-model. The first sub-model outputs the indicator values ​​of the account's N evaluation indicators. The indicator values ​​of the account's N evaluation indicators are input into the second sub-model. The second sub-model outputs the indicator value of the account's comprehensive indicator.

[0284] The fourth sub-model is used to determine the initial parameters of the initial evaluation model before training the initial evaluation model, based on the parameter values ​​of each of the M basic parameters corresponding to the account output by the third sub-model. The fourth sub-model can use methods such as the coefficient of variation method, principal component analysis, and analytic hierarchy process.

[0285] The third sub-model may include a data cleaning module, a parameter determination module, and a normalization module. The data cleaning module is used to clean the account's follower data, work data, and price data. The parameter determination module is used to determine the values ​​of the account's M basic parameters based on the cleaned follower data, work data, and price data. The normalization module is used to normalize the values ​​of the M basic parameters to obtain their parameter values.

[0286] The first and second sub-models are constructed based on AHP. See, for an example, [link to example]. Figure 7 , Figure 7 This is a schematic diagram illustrating the relationship between comprehensive indicators and evaluation indicators, and the influence of evaluation indicators and basic parameters, based on the AHP method disclosed in this application. Figure 7 As shown, the basic parameters related to fan metrics (i.e., the fundamental parameters affecting fan metrics) include: number of fans and fan growth rate; the basic parameters related to update metrics include: update rate and play count; the basic parameters related to viral metrics include: play count and viral rate; the basic parameters related to dissemination metrics include: number of fans, fan growth rate, play count, and stability; and the basic parameters related to cost-effectiveness metrics include CPM and pricing. The evaluation metrics related to comprehensive metrics include: fan metrics, update metrics, viral metrics, dissemination metrics, and cost-effectiveness metrics.

[0287] exist Figure 5In the described data processing method, the initial evaluation model first determines the initial parameters, and then trains the initial evaluation model based on the training data to obtain the target evaluation model. The trained target evaluation model is then used to determine the indicator values ​​for each indicator. This eliminates the need to determine parameters and train the evaluation model every time it is used, thus improving the efficiency of determining indicator values. Based on the account's basic data (i.e., follower data, work data, and price data), the indicator values ​​for N evaluation indicators and a comprehensive indicator of the account can be determined. Therefore, it not only provides the account's basic data but also the indicator values ​​for N evaluation indicators corresponding to different evaluation dimensions determined by the account's basic data, as well as the indicator values ​​for the comprehensive indicator determined based on the indicator values ​​of the N evaluation indicators, enriching the account's data.

[0288] In some embodiments, after the computer device associates and stores the values ​​of N evaluation indicators and the value of the comprehensive indicator with a first account, that is, after storing the values ​​of the N evaluation indicators and the comprehensive indicator of the first account, it can send the values ​​of the N evaluation indicators and the comprehensive indicator of the first account to the client. The client can be any client corresponding to the computer device.

[0289] Accordingly, the client can receive the values ​​of N evaluation indicators and the comprehensive indicator from the first account on the computer device, and then display the values ​​of the N evaluation indicators and the comprehensive indicator of the first account so that users can view them, which can improve the user experience.

[0290] In some embodiments, after the computer device associates and stores the values ​​of N evaluation indicators and the value of a comprehensive indicator with a first account, it can sort the accounts in the first account list according to the value of the first indicator for each account in the first account list, thereby obtaining the ranking of the accounts in the first account list under the first indicator. The first indicator is any one of the N evaluation indicators and the comprehensive indicator. The first account list can be a list of all accounts from the same channel, a list of all accounts from the same region from the same channel, or a list of all accounts from all channels.

[0291] In some embodiments, after obtaining the ranking of accounts in the first account list under the first indicator, the computer device can send the ranking of accounts in the first account list under the first indicator to the client. Correspondingly, the client receives the ranking of accounts in the first account list under the first indicator from the computer device and can then display the ranking of accounts in the first account list under the first indicator, so that the user can select the desired account based on the ranking of accounts in the first account list under the first indicator, thereby improving the user experience.

[0292] In some embodiments, after obtaining the ranking of accounts in the first account list under the first indicator, the computer device can select accounts that meet the first preset ranking criteria from the first account list based on their ranking under the first indicator, and then send the account information of the accounts that meet the first preset ranking criteria from the first account list to the client. Correspondingly, the client receives the account information of the accounts that meet the first preset ranking criteria from the first account list from the computer device, and can then display the accounts that meet the first preset ranking criteria from the first account list, so that the user can view the data of the accounts that meet the first preset ranking criteria, thereby improving the user experience.

[0293] In some embodiments, the first indicator may include L levels, and the first indicator is any one of N evaluation indicators and comprehensive indicators, where L is an integer greater than 1.

[0294] After storing the values ​​of N evaluation indicators and the comprehensive indicator for the first account, the computer equipment can classify the accounts in the second account list according to their indicator values ​​under the first indicator, thus obtaining the corresponding level for each account under the first indicator. Based on the indicator values ​​of each account under the first indicator, the accounts in the second account list corresponding to the same level under the first indicator can be sorted to obtain the ranking of each account in the second account list within the corresponding level for the first indicator. The second account list can be a list of all accounts from the same channel, a list of all accounts from the same region from the same channel, or a list of all accounts from all channels. The second account list can be the same as or different from the first account list.

[0295] For example, assuming the values ​​of each indicator are between 0 and 100, the fan indicator includes five levels: A, B, C, D, and E. The ranges of indicator values ​​corresponding to these five levels are [80, 100), [60, 80), [40, 60), [20, 40), and [0, 20], respectively.

[0296] In some embodiments, the computer device obtains the ranking of each account in the second account list according to a first indicator within a corresponding level, and can send the level and ranking of each account in the second account list under the first indicator to the client. Correspondingly, the client receives the level and ranking of each account in the second account list under the first indicator from the computer device, and then displays the level and ranking of each account in the second account list under the first indicator. This allows users to select the desired accounts based on their rankings under the first indicator, improving the efficiency and accuracy of filtering and enhancing the user experience.

[0297] In some embodiments, after obtaining the level and ranking of each account in the second account list under the first indicator, the computer device can select accounts that meet the second preset ranking criteria from the second account list based on the level and ranking of each account under the first indicator. Then, it can send the account information of the accounts meeting the second preset ranking criteria from the first account list to the client. Correspondingly, the client receives the account information from the computer device's second account list and can then display the accounts meeting the second preset ranking criteria. This allows users to view the data of accounts meeting the second preset ranking criteria, improving the efficiency and accuracy of filtering and enhancing the user experience.

[0298] As can be seen, since the values ​​of each evaluation metric for an account are determined, users can view these values ​​through the client application. Furthermore, because the values ​​of each evaluation metric are fixed, suitable accounts can be selected efficiently and quickly based on these values.

[0299] To better understand the embodiments of this application, the following description will be provided in conjunction with application scenarios. The application scenario below will be illustrated using a game platform as an example.

[0300] See Figure 8 , Figure 8 This is a schematic diagram illustrating an application scenario disclosed in an embodiment of this application. For example... Figure 8 As shown, the gaming platform can periodically obtain the latest follower data, work data, and price data of all influencers on associated platforms, and can store this data. For example, the latest follower data, work data, and price data can be today's data.

[0301] The gaming platform can periodically determine the metric values ​​for various metrics of influencers on affiliated platforms based on their follower, content, and pricing data, and also determine the metric values ​​for influencers on its own platform based on their follower, content, and pricing data. It can then send these metric values ​​to the client. Upon receiving these metric values ​​from both the gaming platform and its own platform, the client can display both sets of metric values.

[0302] For example, see Figure 9 , Figure 9This is a schematic diagram of a client-side interface disclosed in an embodiment of this application. Figure 9 As shown, the client includes multiple interfaces, such as the Homepage, Talent Plaza, Toolbox, and Task Help. The current interface is the Talent Plaza, which includes the Account Platform, followed by information components for associated platforms. The Talent Plaza also includes a "All Talents" component and a "Currently Registered Talents" component (or a component for talents on this platform). When the user selects the "All Talents" component, the client can respond to this selection by displaying the data of all talents, i.e., all talents on this platform and associated platforms. When the user selects the "Currently Registered Talents" component, the client can respond to this selection by displaying the data of all talents on this platform. When the user selects an information component for a specific associated platform following the Account Platform, the client can respond to this selection by displaying the data of all talents on that associated platform.

[0303] like Figure 9 As shown, the current interface displays the influencer's comprehensive metrics, including the number of followers and the follower growth rate over the past 30 days.

[0304] It should be understood that the displayed influencer data can be configured by the user through settings. If the user does not configure settings, the default influencer data will be displayed.

[0305] See Figure 10 , Figure 10 This is a schematic diagram illustrating another application scenario disclosed in the embodiments of this application. For example... Figure 10 As shown, the gaming platform can periodically acquire and store the latest follower, content, and pricing data of all influencers on affiliated platforms. The gaming platform can periodically determine the metric values ​​for various indicators of influencers on affiliated platforms based on their follower, content, and pricing data, and also determine the metric values ​​for various indicators of influencers on its own platform based on their follower, content, and pricing data.

[0306] You can sort influencers on the same platform based on their metrics values ​​to obtain a ranking of influencers across all platforms. Alternatively, you can sort all influencers based on their metrics values ​​to obtain a ranking of all influencers across all platforms. You can also sort influencers of the same category on the same platform based on their metrics values ​​to obtain a ranking of influencers of the same category on the same platform. Finally, you can sort influencers of the same category across all platforms based on their metrics values ​​to obtain a ranking of influencers of the same category across all platforms.

[0307] Then, the system can send the metrics values ​​of influencers from the associated platform, as well as the metrics values ​​and rankings of influencers from its own platform, to the client. After receiving the metrics values ​​and rankings of influencers from the associated platform and influencers from its own platform, the client can display the metrics values ​​and rankings of influencers from both platforms.

[0308] When the client detects a user's action to display a certain influencer's metrics, the client can respond to that action by displaying the influencer's metric values ​​and rankings.

[0309] For example, see Figure 11 , Figure 11 This is a schematic diagram illustrating the indicator values ​​and rankings of various metrics for top performers, as disclosed in an embodiment of this application. For example... Figure 11 As shown, this influencer's overall performance index is 83.3, their follower index is 39, their update index is 96.9, their viral hit index is 91.2, their dissemination index is 81.3, their cost-effectiveness index is 90.6, their ranking among similar influencers is 1907, and their ranking among influencers with the same follower count across the entire network is 1825. Figure 11 As shown, the average value of the fan metric for all online influencers is 33.3, the average value of the update metric is 62.3, the average value of the viral metric is 48.3, the average value of the dissemination metric is 21.7, and the average value of the cost-effectiveness metric is 60.8. Figure 11 As shown, the display can also show the trend of each indicator's value over a period of time; currently, it shows the trend of the propagation indicator. Which indicator's trend is displayed depends on the user's selection.

[0310] The following describes an apparatus embodiment of this application, which can be used to perform the methods described in the above embodiments of this application. For details not disclosed in the apparatus embodiments of this application, please refer to the method embodiments described in the above embodiments of this application.

[0311] See Figure 12 , Figure 12 This is a schematic diagram of a data processing apparatus disclosed in an embodiment of this application. This data processing apparatus can be applied to the aforementioned computer equipment. Figure 12 As shown, the data processing apparatus may include:

[0312] Unit 1201 is used to obtain the first account's follower data, work data, and price data;

[0313] The first determining unit 1202 is used to determine the index values ​​of N evaluation indicators by the first sub-model in the target evaluation model based on the fan data, work data and price data of the first account; N is an integer greater than 1.

[0314] The first determining unit 1202 is also used to determine the index value of the comprehensive index by the second sub-model in the target evaluation model based on the index values ​​of N evaluation indicators;

[0315] Storage unit 1203 is used to associate and store the index values ​​of N evaluation indicators and the index value of the comprehensive indicator with the first account.

[0316] In some embodiments, the acquisition unit is further configured to acquire training data, which includes fan data, work data, price data and tag data of multiple accounts. The tag data of multiple accounts includes the tag index values ​​of N evaluation indicators and the tag index values ​​of the comprehensive indicator of multiple accounts.

[0317] The first determining unit 1202 is also used to determine the index values ​​of N evaluation indicators of the second account based on the fan data, work data and price data of the second account by the first initial sub-model in the initial evaluation model; the second account is any one of multiple accounts;

[0318] The first determining unit 1202 is also used to determine the index value of the comprehensive index of the second account by the second initial sub-model in the initial evaluation model based on the index values ​​of the N evaluation indicators of the second account;

[0319] The data processing apparatus may also include:

[0320] The second determining unit is used to determine the loss value based on the label index values ​​and index values ​​of the N evaluation indicators of the second account, as well as the label index values ​​and index values ​​of the comprehensive indicator.

[0321] The optimization unit is used to optimize the parameters of the initial evaluation model based on the loss value to obtain the target evaluation model.

[0322] In some embodiments, the second determining unit is specifically used for:

[0323] Based on the tag values ​​and values ​​of N indicators of the second account, determine the loss value corresponding to each of the N evaluation indicators;

[0324] The overall loss value is determined based on the tag index value and index value of the comprehensive index of the second account;

[0325] The optimization unit is specifically used for:

[0326] Optimize the weight parameters of the first initial sub-model based on the loss values ​​corresponding to N evaluation indicators;

[0327] The weight parameters of the second initial sub-model are optimized based on the comprehensive loss value.

[0328] In some embodiments, the weight parameters of the first initial sub-model include the weights corresponding to M basic parameters; the weight parameters of the second initial sub-model include the weights corresponding to N evaluation indicators; M is an integer greater than N.

[0329] Unit 1201 is also used to obtain fan data, work data, and price data for K accounts; K is an integer greater than 1.

[0330] The data processing apparatus may also include:

[0331] The third determining unit is used to determine the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts based on the fan data, work data and price data of each account.

[0332] The third determining unit is also used to determine the initial weights corresponding to the M basic parameters based on the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts.

[0333] The third determining unit is also used to determine the indicator values ​​of each of the N evaluation indicators corresponding to each of the K accounts based on the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts and the initial weights corresponding to the M basic parameters.

[0334] The third determining unit is also used to determine the initial weights corresponding to the N evaluation indicators based on the indicator values ​​of each of the N evaluation indicators corresponding to the K accounts.

[0335] An initialization unit is used to initialize the weights corresponding to the M basic parameters in the first initial sub-model to their respective initial weights.

[0336] The initialization unit is also used to initialize the weights corresponding to the N evaluation indicators in the second initial sub-model to their respective initial weights.

[0337] In some embodiments, the initial weights corresponding to the M basic parameters include the initial weights of each of the N evaluation indicators corresponding to each of the M basic parameters.

[0338] The third determining unit determines the initial weights corresponding to the M basic parameters based on the parameter values ​​of each of the M basic parameters for each of the K accounts, including:

[0339] Determine the basic parameters related to the target evaluation index from among the M basic parameters to obtain multiple basic parameters corresponding to the target evaluation index; the target evaluation index is any one of the N evaluation indicators.

[0340] Determine the average value of the target basic parameter for each of the K accounts to obtain the average value of the target basic parameter corresponding to the target evaluation indicator; the target basic parameter is any one of the multiple basic parameters corresponding to the target evaluation indicator.

[0341] Determine the standard deviation of the parameter values ​​of the target basic parameters for each of the K accounts, and obtain the standard deviation of the parameters of the target basic parameters corresponding to the target evaluation indicators;

[0342] Based on the average value and standard deviation of the target basic parameters corresponding to the target evaluation indicators, the initial weights of the target basic parameters corresponding to the target evaluation indicators are determined, thus obtaining the initial weights of each of the M basic parameters corresponding to each of the N evaluation indicators.

[0343] In some embodiments, the third determining unit determines the initial weights corresponding to the N evaluation indicators based on the indicator values ​​of each of the N evaluation indicators corresponding to the K accounts, including:

[0344] Determine the average value of the target evaluation indicator for each of the K accounts to obtain the average value of the target evaluation indicator; the target evaluation indicator is any one of the N evaluation indicators.

[0345] Determine the standard deviation of the indicator values ​​corresponding to the target evaluation indicator for each of the K accounts, and obtain the standard deviation of the indicator corresponding to the target evaluation indicator;

[0346] Based on the average value and standard deviation of the target evaluation indicators, the initial weights of the target evaluation indicators are determined, resulting in the initial weights of N evaluation indicators.

[0347] In some embodiments, the N evaluation metrics include fan metrics, update metrics, hit product metrics, dissemination metrics, and cost-effectiveness metrics.

[0348] In some embodiments, the data processing apparatus may further include:

[0349] The first communication unit is used to send the index values ​​of N evaluation indicators and the index value of the comprehensive indicator of the first account to the client, so that the client can display the index values ​​of the N evaluation indicators and the index value of the comprehensive indicator of the first account.

[0350] In some embodiments, the data processing apparatus may further include:

[0351] The first sorting unit is used to sort the accounts in the first account list according to the indicator value of the first indicator of each account in the first account list, so as to obtain the ranking of the accounts in the first account list under the first indicator. The first indicator is any one of N evaluation indicators and comprehensive indicators.

[0352] The second communication unit is used to send the ranking of the accounts in the first account list under the first indicator to the client, so that the client can display the ranking of the accounts in the first account list under the first indicator.

[0353] In some embodiments, the first indicator includes L levels, and the first indicator is any one of N evaluation indicators and a comprehensive indicator, where L is an integer greater than 1. The data processing device may further include:

[0354] The classification unit is used to classify the accounts in the second account list according to the indicator value of each account in the first indicator, so as to obtain the corresponding level of each account in the second account list under the first indicator.

[0355] The second sorting unit is used to sort the accounts in the second account list that are at the same level under the first indicator according to the indicator value of each account under the first indicator, so as to obtain the ranking of each account in the second account list in the corresponding level under the first indicator.

[0356] The third communication unit is used to send the level and ranking of each account in the second account list under the first indicator to the client, so that the client can display the level and ranking of each account in the second account list under the first indicator.

[0357] In some embodiments, after the first sorting unit sorts the accounts in the first account list according to the index value of the first index of each account in the first account list, and obtains the ranking of the accounts in the first account list under the first index, the data processing device may further include:

[0358] The selection unit is used to select accounts that meet the preset ranking conditions from the first account list based on the ranking of the accounts in the first account list under the first indicator.

[0359] The fourth communication unit is used to send account information of accounts that meet the preset ranking conditions in the first account list to the client.

[0360] The specific working processes of the acquisition unit 1201, the first determination unit 1202, the storage unit 1203, the second determination unit, the optimization unit, the third determination unit, the initialization unit, the first communication unit, the first sorting unit, the second communication unit, the classification unit, the second sorting unit, the third communication unit, the selection unit, and the fourth communication unit can be referred to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0361] See Figure 13 , Figure 13 This is a schematic diagram of the structure of a computer system of a computer device disclosed in an embodiment of this application.

[0362] It should be noted that, Figure 13 The computer system of the computer device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application. The computer device can be used to perform the data processing methods provided in this application.

[0363] like Figure 13 As shown, the computer system includes a Central Processing Unit (CPU) 1301, which can perform various appropriate actions and processes, such as executing the methods described in the above embodiments, based on programs stored in Read-Only Memory (ROM) 1302 or programs loaded from storage portion 1308 into Random Access Memory (RAM) 1303. The RAM 1303 also stores various programs and data required for system operation. The CPU 1301, ROM 1302, and RAM 1303 are interconnected via a bus 1304. An Input / Output (I / O) interface 1305 is also connected to the bus 1304.

[0364] The following components are connected to I / O interface 1305: an input section 1306 including a keyboard, mouse, etc.; an output section 1307 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 1308 including a hard disk, etc.; and a communication section 1309 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 1309 performs communication processing via a network such as the Internet. A drive 1310 is also connected to I / O interface 1305 as needed. Removable media 1311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 1310 as needed so that computer programs read from them can be installed into storage section 1308 as needed.

[0365] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 1309, and / or installed from removable medium 1311. When the computer program is executed by central processing unit (CPU) 1301, it performs various functions defined in the system of this application.

[0366] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such transmitted data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0367] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0368] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0369] In another aspect, this application also provides a computer-readable storage medium, which may be included in the computer device described in the above embodiments; or it may exist independently and not assembled into the computer device. The aforementioned computer-readable storage medium carries computer-readable instructions that, when executed by a processor, implement the methods in any of the above embodiments.

[0370] In the embodiments of this application, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal. It can be implemented wholly or partially using software, hardware (e.g., processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that functions as a whole.

[0371] According to one aspect of the embodiments of this application, a computer program product or computer program is provided, the computer program product or computer program including 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 the processor executes the computer instructions, causing the computer device to perform the methods of any of the above embodiments.

[0372] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0373] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the method according to the embodiments of this application.

[0374] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the embodiments disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0375] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A data processing method, characterized in that, include: Obtain the first account's follower data, work data, and price data; The first sub-model in the target evaluation model determines the values ​​of N evaluation indicators based on the fan data, the work data, and the price data; N is an integer greater than 1. The second sub-model in the target evaluation model determines the value of the comprehensive indicator based on the values ​​of the N evaluation indicators. The values ​​of the N evaluation indicators and the value of the comprehensive indicator are associated with and stored with the first account.

2. The method according to claim 1, characterized in that, The method further includes: Acquire training data, which includes follower data, work data, price data, and tag data of multiple accounts. The tag data of the multiple accounts includes the tag index values ​​of N evaluation indicators and the tag index values ​​of the comprehensive indicator of the multiple accounts. The first initial sub-model in the initial evaluation model determines the index values ​​of N evaluation indicators for the second account based on the fan data, work data, and price data of the second account; the second account can be any one of the multiple accounts. The second initial sub-model in the initial evaluation model determines the comprehensive index value of the second account based on the index values ​​of the N evaluation indicators of the second account; The loss value is determined based on the label and indicator values ​​of the N evaluation indicators of the second account, as well as the label and indicator values ​​of the comprehensive indicator. The parameters of the initial evaluation model are optimized based on the loss value to obtain the target evaluation model.

3. The method according to claim 2, characterized in that, The step of determining the loss value based on the label and indicator values ​​of the N evaluation indicators of the second account, and the label and indicator values ​​of the comprehensive indicator, includes: Based on the tag value and value of N indicators of the second account, determine the loss value corresponding to each of the N evaluation indicators; The overall loss value is determined based on the tag index value and index value of the comprehensive index of the second account; The optimization of the parameters of the initial evaluation model based on the loss value includes: The weight parameters of the first initial sub-model are optimized based on the loss values ​​corresponding to the N evaluation indicators. The weight parameters of the second initial sub-model are optimized based on the comprehensive loss value.

4. The method according to claim 3, characterized in that, The weight parameters of the first initial sub-model include the weights corresponding to M basic parameters; The weight parameters of the second initial sub-model include the weights corresponding to the N evaluation indicators respectively; M is an integer greater than N; The method further includes: Retrieve follower data, artwork data, and price data for K accounts; K is an integer greater than 1. Based on the fan data, work data, and price data of each of the K accounts, determine the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts; Based on the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts, determine the initial weights corresponding to the M basic parameters respectively; Based on the parameter values ​​of each of the M basic parameters corresponding to each of the K accounts, and the initial weights corresponding to the M basic parameters, the indicator values ​​of each of the N evaluation indicators corresponding to each of the K accounts are determined. Based on the index values ​​of each of the N evaluation indicators corresponding to the K accounts, the initial weights corresponding to the N evaluation indicators are determined. The weights corresponding to the M basic parameters in the first initial sub-model are initialized to their respective initial weights; The weights corresponding to the N evaluation indicators in the second initial sub-model are initialized to their respective initial weights.

5. The method according to claim 4, characterized in that, The initial weights corresponding to the M basic parameters include the initial weights of each of the M basic parameters corresponding to each of the N evaluation indicators. The step of determining the initial weights corresponding to the M basic parameters based on the parameter values ​​of each of the M basic parameters for each of the K accounts includes: Determine the basic parameters related to the target evaluation index from the M basic parameters to obtain multiple basic parameters corresponding to the target evaluation index; the target evaluation index is any one of the N evaluation indicators. The average value of the target basic parameter corresponding to each of the K accounts is determined to obtain the average value of the target basic parameter corresponding to the target evaluation indicator; the target basic parameter is any one of the multiple basic parameters corresponding to the target evaluation indicator. Determine the standard deviation of the parameter values ​​of the target basic parameter for each of the K accounts to obtain the standard deviation of the parameter values ​​of the target evaluation index corresponding to the target basic parameter; Based on the average value and standard deviation of the target basic parameters corresponding to the target evaluation index, the initial weights of the target basic parameters corresponding to the target evaluation index are determined, thereby obtaining the initial weights of each of the M basic parameters corresponding to each of the N evaluation indicators.

6. The method according to claim 4, characterized in that, The step of determining the initial weights corresponding to the N evaluation indicators based on the indicator values ​​of the N evaluation indicators corresponding to the K accounts includes: The average value of the target evaluation indicator corresponding to each of the K accounts is determined to obtain the average value of the target evaluation indicator; the target evaluation indicator is any one of the N evaluation indicators. Determine the standard deviation of the indicator values ​​corresponding to the target evaluation indicator for each of the K accounts, and obtain the standard deviation of the indicator corresponding to the target evaluation indicator; Based on the average value and standard deviation of the target evaluation index, the initial weights corresponding to the target evaluation index are determined, and the initial weights corresponding to the N evaluation indicators are obtained.

7. The method according to claim 1, characterized in that, The N evaluation metrics include fan metrics, update metrics, hit product metrics, dissemination metrics, and cost-effectiveness metrics.

8. The method according to any one of claims 1-7, characterized in that, The method further includes: Send the values ​​of N evaluation indicators and the value of the comprehensive indicator of the first account to the client so that the client can display the values ​​of the N evaluation indicators and the value of the comprehensive indicator of the first account.

9. The method according to any one of claims 1-7, characterized in that, The method further includes: Based on the indicator value of the first indicator of each account in the first account list, the accounts in the first account list are sorted to obtain the ranking of the accounts in the first account list under the first indicator. The first indicator is any one of the N evaluation indicators and the comprehensive indicator. Send the ranking of the accounts in the first account list under the first metric to the client, so that the client can display the ranking of the accounts in the first account list under the first metric.

10. The method according to any one of claims 1-7, characterized in that, The first indicator includes L levels, and the first indicator is any one of the N evaluation indicators and the comprehensive indicator, where L is an integer greater than 1. The method further includes: Based on the indicator value of each account in the second account list under the first indicator, the accounts in the second account list are classified into levels to obtain the level corresponding to each account in the second account list under the first indicator; Based on the indicator value of each account under the first indicator, the accounts in the second account list that correspond to the same level under the first indicator are sorted to obtain the ranking of each account in the second account list in the corresponding level of the first indicator. The system sends the level and ranking of each account in the second account list under the first indicator to the client, so that the client can display the level and ranking of each account in the second account list under the first indicator.

11. The method according to claim 9, characterized in that, After sorting the accounts in the first account list according to the indicator value of the first indicator for each account in the first account list, and obtaining the ranking of the accounts in the first account list under the first indicator, the method further includes: Based on the ranking of the accounts in the first account list under the first indicator, select accounts that meet the preset ranking conditions from the first account list; Send the account information of accounts in the first account list that meet the preset ranking conditions to the client.

12. A data processing apparatus, characterized in that, include: The acquisition unit is used to acquire the first account's fan data, work data, and price data; The first determining unit is used to determine the index values ​​of N evaluation indicators by the first sub-model in the target evaluation model based on the fan data, the work data, and the price data; N is an integer greater than 1. The first determining unit is further configured to determine the index value of the comprehensive index by the second sub-model in the target evaluation model based on the index values ​​of the N evaluation indicators; The storage unit is used to associate and store the index values ​​of the N evaluation indicators and the index value of the comprehensive indicator with the first account.

13. A computer device, characterized in that, include: processor; A memory storing computer-readable instructions that, when executed by the processor, implement the method as described in any one of claims 1-11.

14. A computer-readable storage medium storing computer-readable instructions thereon, characterized in that, When the computer-readable instructions are executed by a processor, the method as described in any one of claims 1-11 is implemented.

15. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, the method as described in any one of claims 1-11 is implemented.

Images