Channel behavior quality evaluation method, channel quality evaluation method, and related device

Abstract

Embodiments of the present application provide a channel behavior quality evaluation method, a channel quality evaluation method and related equipment. The channel behavior quality evaluation method comprises: obtaining device behavior similarity between a plurality of devices corresponding to each channel according to channel behavior data of each channel respectively; determining a channel cohesion value and a channel diversity value corresponding to each channel according to each device behavior similarity corresponding to each channel; and determining a channel behavior quality score corresponding to each channel according to a channel benchmark value, the channel cohesion value and the channel diversity value corresponding to each channel. The technical solution of the embodiments of the present application enriches the buying channel quality evaluation system by introducing behavior data and multi-dimensional evaluation indicators and other behavior-related factors, maximizes the accuracy of the evaluation, optimizes the channel selection process, and thus improves the scientificity and efficiency of the advertisement placement and promotion effect.

Description

Technical Field

[0001] This application relates to the fields of computer and communication technology, and more specifically, to a channel behavior quality assessment method, a channel quality assessment method, and related equipment. Background Technology

[0002] Current evaluation systems for assessing the quality of user acquisition channels primarily measure three main dimensions: new user acquisition, retention, and revenue / expenses. However, these systems fail to adequately consider the diversity and consistency of traffic within each channel and easily overlook the interference of fraudulent traffic. Therefore, accurately evaluating the quality of user acquisition channels, identifying fraudulent traffic, and selecting high-quality traffic has become a significant technical challenge. Summary of the Invention

[0003] The embodiments of this application provide a channel behavior quality assessment method, a channel quality assessment method, and related equipment, which can at least to a certain extent achieve accurate assessment of the quality of traffic acquisition channels, identify fake traffic, and select high-quality traffic.

[0004] Other features and advantages of this application will become apparent from the following detailed description, or may be learned in part from practice of this application.

[0005] According to one aspect of the embodiments of this application, a method for evaluating channel behavior quality is provided, comprising: obtaining device behavior similarity among multiple devices corresponding to each channel based on channel behavior data for each channel; determining channel cohesion value and channel diversity value for each channel based on the device behavior similarity for each channel; and determining channel behavior quality score for each channel based on the channel benchmark value, channel cohesion value, and channel diversity value for each channel.

[0006] In some embodiments of this application, the step of obtaining the device behavior similarity among multiple devices corresponding to each channel based on the channel behavior data of each channel specifically includes: extracting device behavior representations from the channel behavior data of each channel using a behavior representation extraction model to obtain device behavior chains; performing same-dimensional processing on each device behavior chain corresponding to each channel to obtain corresponding same-dimensional device behavior chains; and comparing each same-dimensional device behavior chain corresponding to each channel to obtain the device behavior similarity among the devices.

[0007] In some embodiments of this application, the step of performing same-dimensional processing on each device behavior chain corresponding to each channel to obtain the corresponding same-dimensional device behavior chain specifically includes: performing completion processing on each device behavior chain corresponding to each channel to obtain a completed device behavior chain; and performing same-position alignment processing on each device behavior chain corresponding to each channel to obtain the same-dimensional device behavior chain.

[0008] In some embodiments of this application, the behavior representation extraction model includes an encoder and a decoder, and the method further includes: acquiring a device behavior data sequence sample set, the device behavior data sequence sample set containing multiple device behavior data sequence samples, each of the device behavior data sequence samples being labeled with a corresponding behavior representation sequence; inputting the device behavior data sequence samples into the encoder to obtain a behavior sequence representation; inputting the behavior sequence representation into the decoder to obtain a behavior representation vector; determining a loss function based on the finally obtained behavior representation vector and the labeled behavior representation sequence; and updating the parameters of the encoder and the decoder based on the loss function until the loss function converges.

[0009] In some embodiments of this application, determining the channel cohesion value and channel diversity value corresponding to each channel based on the similarity of the device behaviors corresponding to each channel specifically includes: taking the mean of the similarity of the device behaviors corresponding to each channel to obtain the channel cohesion value; and taking the standard deviation of the similarity of the device behaviors corresponding to each channel to obtain the channel diversity value.

[0010] In some embodiments of this application, the channel benchmark value includes a channel cohesion benchmark value and a channel diversity benchmark value. The step of determining the channel behavior quality score for each channel based on its corresponding channel benchmark value, channel cohesion value, and channel diversity value specifically includes: determining the channel behavior cohesion score for each channel based on its corresponding channel cohesion benchmark value and channel cohesion value, wherein the channel cohesion benchmark value is a cohesion value determined from organic traffic; determining the channel behavior diversity score for each channel based on its corresponding channel diversity benchmark value and channel diversity value, wherein the channel diversity benchmark value is a diversity value determined from organic traffic; and obtaining the channel behavior quality score for each channel based on its corresponding channel behavior cohesion score and channel behavior diversity score.

[0011] In some embodiments of this application, determining the channel behavior cohesion score for each channel based on the channel cohesion benchmark value and the channel cohesion value specifically includes: comparing the channel cohesion benchmark value and the channel cohesion value for each channel; if the channel cohesion value for a given channel is smaller than the channel cohesion benchmark value, then the channel cohesion value is used as the cohesion behavior amplitude difference; if the channel cohesion value for a given channel is larger than the channel cohesion benchmark value, then the difference between the unit value and the channel cohesion value is used as the cohesion behavior amplitude difference; determining the maximum and minimum cohesion behavior amplitude differences among the cohesion behavior amplitude differences for each channel; and obtaining the channel behavior cohesion score based on the maximum cohesion behavior amplitude difference, the minimum cohesion behavior amplitude difference, and the channel cohesion value for each channel.

[0012] In some embodiments of this application, determining the channel behavior diversity score for each channel based on the channel diversity benchmark value and the channel diversity value for each channel specifically includes: comparing the channel diversity benchmark value and the channel diversity value for each channel; if the channel diversity value for a channel is smaller than the channel diversity benchmark value, then the channel diversity value is used as the diversity behavior amplitude difference; if the channel diversity value for a channel is larger than the channel diversity benchmark value, then the difference between the unit value and the channel diversity value is used as the diversity behavior amplitude difference; determining the maximum diversity behavior amplitude difference and the minimum diversity behavior amplitude difference among the diversity behavior amplitude differences for each channel; and obtaining the channel behavior diversity score based on the maximum diversity behavior amplitude difference, the minimum diversity behavior amplitude difference, and the channel diversity value for each channel.

[0013] In some embodiments of this application, obtaining the channel behavior quality score corresponding to each channel based on the channel behavior cohesion score and the channel behavior diversity score specifically includes: weighted summing of the channel behavior cohesion score and the channel behavior diversity score corresponding to each channel to obtain the channel behavior quality score corresponding to each channel.

[0014] According to one aspect of the embodiments of this application, a channel quality assessment method is provided. The channel behavior quality assessment method includes: acquiring channel assessment data, the channel assessment data including channel new user data, channel retention data, channel revenue data, and channel behavior data; determining a channel new user quality score based on the channel new user data; determining a channel retention quality score based on the channel retention data; determining a channel revenue quality score based on the channel revenue data; determining a channel behavior quality score based on the channel behavior data using the channel behavior quality assessment method described above; and determining a channel quality assessment score based on the channel new user quality score, the channel retention quality score, the channel revenue quality score, and the channel behavior quality score.

[0015] In some embodiments of this application, determining the channel addition quality score based on the channel addition data specifically includes: determining the new channel addition score and the existing channel addition score based on the new channel addition data; and determining the channel addition quality score based on the new channel addition score and the existing channel addition score.

[0016] In some embodiments of this application, determining a channel retention quality score based on the channel retention data specifically includes: determining the next-day retention rate, weekly retention rate, and long-term retention rate based on the channel retention data; and determining the channel retention quality score based on the next-day retention rate, the weekly retention rate, and the long-term retention rate.

[0017] In some embodiments of this application, determining the channel revenue quality score based on the channel revenue data specifically includes: determining active device revenue, channel revenue, and channel cost based on the channel revenue data; and determining the channel revenue quality score based on the active device revenue, the channel revenue, and the channel cost.

[0018] In some embodiments of this application, determining the channel quality assessment score based on the channel new user quality score, the channel retention quality score, the channel revenue quality score, and the channel behavior quality score specifically includes: performing a weighted summation of the channel new user quality score, the channel retention quality score, the channel revenue quality score, and the channel behavior quality score to obtain the channel quality assessment score.

[0019] According to one aspect of the embodiments of this application, a channel behavior quality assessment device is provided, the channel behavior quality assessment device comprising: a device similarity determination module, configured to obtain device behavior similarity among multiple devices corresponding to each channel based on channel behavior data for each channel; a cohesion and diversity determination module, configured to determine channel cohesion value and channel diversity value corresponding to each channel based on the device behavior similarity corresponding to each channel; and a behavior score determination module, configured to determine channel behavior quality score corresponding to each channel based on channel benchmark value, channel cohesion value, and channel diversity value corresponding to each channel.

[0020] In some embodiments of this application, the device similarity determination module specifically includes: a behavior representation extraction submodule, used to extract device behavior representations from the channel behavior data of each channel using a behavior representation extraction model to obtain device behavior chains; a device same-dimensional processing submodule, used to perform same-dimensional processing on each device behavior chain corresponding to each channel to obtain corresponding same-dimensional device behavior chains; and a device similarity determination submodule, used to compare each same-dimensional device behavior chain corresponding to each channel to obtain the device behavior similarity between devices.

[0021] In some embodiments of this application, the device same-dimensional processing submodule specifically includes: a behavior chain completion unit, used to complete each device behavior chain corresponding to each channel to obtain a completed device behavior chain; and a same-position alignment unit, used to perform same-position alignment processing on each device behavior chain corresponding to each channel to obtain a same-dimensional device behavior chain.

[0022] In some embodiments of this application, the behavior representation extraction model includes an encoder and a decoder, and the channel behavior quality assessment device further includes: a sample acquisition module for acquiring a set of device behavior data sequence samples, wherein the set of device behavior data sequence samples contains multiple device behavior data sequence samples, and each device behavior data sequence sample is labeled with a corresponding behavior representation sequence; a sample encoding module for inputting the device behavior data sequence samples into the encoder to obtain behavior sequence representations; a sample decoding module for inputting the behavior sequence representations into the decoder to obtain behavior representation vectors; a loss function module for determining a loss function based on the finally obtained behavior representation vectors and the labeled behavior representation sequences; and a parameter update module for updating the parameters of the encoder and the decoder based on the loss function until the loss function converges.

[0023] In some embodiments of this application, the cohesion diversity determination module specifically includes: a channel cohesion value submodule, used to take the mean of the similarity of the behavior of each device corresponding to each channel to obtain a channel cohesion value; and a channel diversity value submodule, used to take the standard deviation of the similarity of the behavior of each device corresponding to each channel to obtain a channel diversity value.

[0024] In some embodiments of this application, the channel benchmark value includes a channel cohesion benchmark value and a channel diversity benchmark value. The behavior score determination module specifically includes: a behavior cohesion numerator module, used to determine the channel behavior cohesion score corresponding to each channel based on the channel cohesion benchmark value and the channel cohesion value corresponding to each channel, wherein the channel cohesion benchmark value is a cohesion value determined from natural traffic; a behavior diversity numerator module, used to determine the channel behavior diversity score corresponding to each channel based on the channel diversity benchmark value and the channel diversity value corresponding to each channel, wherein the channel diversity benchmark value is a diversity value determined from natural traffic; and a channel behavior quality numerator module, used to obtain the channel behavior quality score corresponding to each channel based on the channel behavior cohesion score and the channel behavior diversity score corresponding to each channel.

[0025] In some embodiments of this application, the behavioral cohesion module specifically includes: a cohesion benchmark comparison unit, used to compare the channel cohesion benchmark value with the channel cohesion value corresponding to each channel; a first cohesion amplitude unit, used to use the channel cohesion value as the cohesion behavioral amplitude difference if the channel cohesion value corresponding to the channel is smaller than the channel cohesion benchmark value; a second cohesion amplitude unit, used to use the difference between the unit value and the channel cohesion value as the cohesion behavioral amplitude difference if the channel cohesion value corresponding to the channel is larger than the channel cohesion benchmark value; a cohesion amplitude extreme value unit, used to determine the maximum and minimum cohesion behavioral amplitude differences among the cohesion behavioral amplitude differences corresponding to each channel; and a cohesion score calculation unit, used to obtain the channel behavioral cohesion score based on the maximum cohesion behavioral amplitude difference, the minimum cohesion behavioral amplitude difference, and the channel cohesion value corresponding to each channel.

[0026] In some embodiments of this application, the behavioral diversity module specifically includes: a diversity benchmark comparison unit, used to compare the channel diversity benchmark value with the channel diversity value corresponding to each channel; a first diversity amplitude unit, used to use the channel diversity value as the diversity behavioral amplitude difference if the channel diversity value corresponding to the channel is smaller than the channel diversity benchmark value; a second diversity amplitude unit, used to use the difference between the unit value and the channel diversity value as the diversity behavioral amplitude difference if the channel diversity value corresponding to the channel is larger than the channel diversity benchmark value; a diversity amplitude extreme value unit, used to determine the maximum and minimum diversity behavioral amplitude differences among the diversity behavioral amplitude differences corresponding to each channel; and a diversity score calculation unit, used to obtain a channel behavioral diversity score based on the maximum diversity behavioral amplitude difference, the minimum diversity behavioral amplitude difference, and the channel diversity value corresponding to each channel.

[0027] In some embodiments of this application, the channel behavior quality module specifically includes: a channel behavior quality scoring unit, used to perform a weighted summation of the channel behavior cohesion score and the channel behavior diversity score corresponding to each channel to obtain the channel behavior quality score corresponding to each channel.

[0028] According to one aspect of the embodiments of this application, a channel quality assessment system is provided, the channel quality assessment system comprising: a channel quality data acquisition device, configured to acquire channel assessment data, the channel assessment data including channel new addition data, channel retention data, channel revenue data, and channel behavior data; a channel new addition quality assessment device, configured to determine a channel new addition quality score based on the channel new addition data; a channel retention quality assessment device, configured to determine a channel retention quality score based on the channel retention data; a channel revenue quality assessment device, configured to determine a channel revenue quality score based on the channel revenue data using the channel behavior quality assessment method described above; and a channel quality assessment scoring device, configured to determine a channel quality assessment score based on the channel new addition quality score, the channel retention quality score, the channel revenue quality score, and the channel behavior quality score.

[0029] In some embodiments of this application, the channel quality data acquisition device specifically includes: a channel addition score module, used to determine the new channel addition score and the existing channel addition score respectively based on the new channel data; and an addition quality score module, used to determine the new channel quality score based on the new channel addition score and the existing channel addition score.

[0030] In some embodiments of this application, the channel retention quality assessment device specifically includes: a retention parameter determination module, used to determine the next-day retention rate, weekly retention rate, and long-term retention rate based on the channel retention data; and a retention quality score module, used to determine the channel retention quality score based on the next-day retention rate, the weekly retention rate, and the long-term retention rate.

[0031] In some embodiments of this application, the channel revenue quality assessment device specifically includes: a revenue cost determination module, used to determine active device revenue, channel revenue, and channel cost based on the channel revenue data; and a revenue quality scoring module, used to determine a channel revenue quality score based on the active device revenue, the channel revenue, and the channel cost.

[0032] In some embodiments of this application, the channel quality assessment scoring device specifically includes: a channel quality assessment scoring module, used to perform a weighted summation of the channel new user quality score, the channel retention quality score, the channel revenue quality score, and the channel behavior quality score to obtain a channel quality assessment score.

[0033] According to one aspect of the embodiments of this application, a computer program product is provided, including one or more computer programs, which, when executed by one or more processors, implement the channel behavior quality assessment method as described in the above embodiments or implement the channel quality assessment method as described in the above embodiments.

[0034] According to one aspect of the embodiments of this application, a computer-readable medium is provided having a computer program stored thereon, which, when executed by a processor, implements the channel behavior quality assessment method as described in the above embodiments or implements the channel quality assessment method as described in the above embodiments.

[0035] According to one aspect of the embodiments of this application, an electronic device is provided, including: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement the channel behavior quality assessment method as described in the above embodiments or to implement the channel quality assessment method as described in the above embodiments.

[0036] In some embodiments of this application, the technical solutions provide for evaluating the consistency of traffic within a channel by analyzing device behavior similarity, thus helping to identify fraudulent traffic. Simultaneously, by combining behavioral consistency and behavioral differences with benchmark comparisons, a comprehensive quality score for each channel is derived. This allows for a more comprehensive analysis of channel traffic quality, aiding in optimization decisions and effectively addressing the technical issues that traditional evaluation methods fail to address. By introducing behavioral data and multi-dimensional evaluation indicators, the system enriches the quality evaluation of user acquisition channels, maximizing evaluation accuracy, optimizing the channel selection process, and ultimately improving the scientific rigor and efficiency of advertising placement and promotion.

[0037] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0038] 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. In the drawings:

[0039] Figure 1 A schematic diagram of an exemplary system architecture to which the technical solutions of the embodiments of this application can be applied is shown.

[0040] Figure 2 A flowchart illustrating a channel behavior quality assessment method provided in an embodiment of this application is shown.

[0041] Figure 3 It shows that according to Figure 2 A flowchart illustrating a specific implementation of step S510 in the channel behavior quality assessment method shown in the corresponding embodiment.

[0042] Figure 4 It shows that according to Figure 2 A flowchart illustrating a specific implementation of step S530 in the channel behavior quality assessment method shown in the corresponding embodiment.

[0043] Figure 5 A flowchart illustrating a channel quality assessment method provided in an embodiment of this application is shown.

[0044] Figure 6 A schematic diagram of the structure of a channel behavior quality assessment device provided in an embodiment of this application is shown.

[0045] Figure 7 A schematic diagram of the structure of a channel quality assessment device provided in an embodiment of this application is shown.

[0046] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown. Detailed Implementation

[0047] 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.

[0048] 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.

[0049] 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 modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0050] 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.

[0051] Figure 1 A schematic diagram of an exemplary system architecture to which the technical solutions of the embodiments of this application can be applied is shown.

[0052] like Figure 1 As shown, the system architecture may include terminal devices (such as...) Figure 1 The device shown includes one or more of a smartphone 101, tablet 102, and portable computer 103 (which could also be a desktop computer, etc.), a network 104, and a server 105. The network 104 serves as a medium for providing a communication link between the terminal device and the server 105. The network 104 can include various connection types, such as wired communication links, wireless communication links, etc.

[0053] It should be understood that Figure 1The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, there can be any number of terminal devices, networks, and servers. For example, server 105 could be a server cluster composed of multiple servers.

[0054] Users can use terminal devices to interact with server 105 via network 104 to receive or send messages, etc. Server 105 can be a server that provides various services. For example, a user can use terminal device 103 (or terminal device 101 or 102) to upload channel behavior data of various channels to server 105. Server 105 can obtain the device behavior similarity between multiple devices corresponding to each channel based on the channel behavior data of each channel; determine the channel cohesion value and channel diversity value corresponding to each channel based on the device behavior similarity corresponding to each channel; and determine the channel behavior quality score corresponding to each channel based on the channel benchmark value, channel cohesion value, and channel diversity value corresponding to each channel.

[0055] It should be noted that the channel behavior quality assessment method provided in this application embodiment is generally executed by server 105, and correspondingly, the channel behavior quality assessment device is generally set in server 105. However, in other embodiments of this application, the terminal device may also have similar functions to the server, thereby executing the channel behavior quality assessment scheme provided in this application embodiment.

[0056] The implementation details of the technical solutions in the embodiments of this application are described in detail below:

[0057] Figure 2 A flowchart of a channel behavior quality assessment method according to an embodiment of this application is shown. This channel behavior quality assessment method can be executed by a server, which may be... Figure 1 The server shown. (Refer to...) Figure 2 As shown, the channel behavior quality assessment method includes at least the following:

[0058] S510, based on the channel behavior data of each channel, obtain the device behavior similarity among multiple devices corresponding to the channel.

[0059] S520, based on the similarity of the behaviors of each device corresponding to each channel, determine the channel cohesion value and channel diversity value corresponding to each channel.

[0060] S530 determines the channel behavior quality score for each channel based on the channel benchmark value, channel cohesion value, and channel diversity value corresponding to each channel.

[0061] In the embodiments of this application, the consistency of traffic within a channel is assessed by analyzing the similarity of device behavior, which helps identify fraudulent traffic. Simultaneously, by combining behavioral consistency and behavioral differences with benchmark comparisons, a comprehensive quality score is derived for each channel. This allows for a more comprehensive analysis of channel traffic quality, aiding in optimization decisions and effectively addressing the technical limitations of traditional evaluation methods. By introducing behavioral data and multi-dimensional evaluation indicators, the quality evaluation system for user acquisition channels is enriched, maximizing the accuracy of the evaluation, optimizing the channel selection process, and ultimately improving the scientific rigor and efficiency of advertising placement and promotion.

[0062] In S510, by analyzing device behavior similarity, it is possible to initially identify whether there are large-scale fraudulent behavior patterns (e.g., automated, non-genuine user behavior) within a channel. Channels with high device behavior similarity often indicate that the traffic from that channel is more authentic, stable, and of higher quality.

[0063] Specifically, in some embodiments, the specific implementation of step S100 can be found in [reference needed]. Figure 3 . Figure 3 It is based on Figure 2 According to the detailed description of step S100 in the channel behavior quality assessment method shown in the corresponding embodiment, step S100 in the channel behavior quality assessment method may include the following steps:

[0064] S512 extracts device behavior representations from the channel behavior data of each channel using a behavior representation extraction model to obtain the device behavior chain.

[0065] S514, perform same-dimensional processing on each device behavior chain corresponding to each channel to obtain the corresponding same-dimensional device behavior chain.

[0066] S516 compares the behavior chains of each device in the same dimension corresponding to each channel to obtain the device behavior similarity between devices.

[0067] In this embodiment, firstly, the behavior representation extraction model can accurately extract device behavior features, avoiding the complexity and noise interference caused by directly processing raw data. Through the extraction of behavior features and the construction of device behavior chains, it can effectively reflect the interaction process of devices in the channel, thereby improving the accuracy of behavior similarity calculation. Secondly, same-dimensional processing ensures the consistency of device behavior data from different channels in terms of dimensions. By eliminating dimensional differences between different channels and devices, subsequent similarity calculations can eliminate the interference caused by data differences between channels, making the comparison of behavior similarity between different devices more reasonable and fair. Finally, through the calculation of behavior similarity, the quality of traffic within the channel can be accurately assessed, and fraudulent traffic can be effectively identified.

[0068] Fake traffic often manifests as random or abnormal device behavior. Therefore, by comparing device behavior chains and calculating similarity, devices exhibiting abnormal behavior can be quickly identified, thereby filtering out invalid traffic.

[0069] This embodiment can handle data differences from different channels and devices, making it applicable to a wider range of scenarios. Whether it's a channel with significant differences in data dimensions or a situation with complex device behavior, efficient and accurate evaluation can be achieved through same-dimensional processing and behavior chain comparison techniques.

[0070] In S512, the behavior representation extraction model effectively extracts meaningful behavioral features from complex raw data, constructing the device's behavior chain. The behavior chain is a tool that accurately represents the device's behavioral characteristics, providing a clear basis for subsequent behavior similarity calculations.

[0071] The aforementioned behavior representation extraction model includes an encoder and a decoder, and its specific training method may include the following steps:

[0072] Obtain a set of device behavior data sequence samples, which contains multiple device behavior data sequence samples, each of which is labeled with a corresponding behavior representation sequence.

[0073] The device behavior data sequence sample is input into the encoder to obtain the behavior sequence representation.

[0074] The behavior sequence representation is input into the decoder to obtain the behavior representation vector.

[0075] Based on the final obtained behavior representation vector and the labeled behavior representation sequence, the loss function is determined.

[0076] The encoder and decoder are updated based on the loss function until the loss function converges.

[0077] In this embodiment, multiple device behavior data sequence samples are first acquired, and each sample is labeled with a corresponding behavior representation sequence. The behavior data sequence samples typically contain a series of device behavior records within a certain time period. These records may include device operation behaviors, timestamps, behavior types, etc. Each device behavior data sequence is labeled with a corresponding "behavior representation sequence," which reflects the characteristics of the device behavior. The device behavior data sequence samples are input into an encoder, which uses a deep learning model to transform the original device behavior data sequence into a low-dimensional behavior sequence representation. The encoder typically employs deep neural network architectures such as LSTM, GRU, or Transformer, which can effectively capture the temporal features and long-short-term dependencies of the data. The behavior sequence representation processed by the encoder is input into a decoder, whose task is to further transform the behavior sequence representation output by the encoder into a final behavior representation vector. The decoder is symmetrical to the encoder; common architectures include Transformer decoders or LSTM decoders. Its purpose is to further map the compressed representation into a vector that reflects the behavior features. The loss function is designed based on the difference between the final behavior representation vector and the labeled behavior representation sequence. Mean squared error (MSE), cross-entropy loss, or other loss functions suitable for sequence generation are typically used to measure the model's prediction error by calculating the difference between the output behavioral representation and the target behavioral representation sequence. The encoder and decoder parameters are updated using the backpropagation algorithm based on the gradient calculated from the loss function. This parameter update process continues until the loss function converges, i.e., the difference between the model's output and the labeled behavioral representation sequence reaches its minimum.

[0078] This embodiment, through joint training of the encoder and decoder, can extract a more concise and effective representation vector from complex raw data of device behavior data, thereby avoiding the high dimensionality and sparsity problems of the original data. The design of the loss function and the optimization of gradient descent enable the model to continuously learn the patterns in the device behavior data and ultimately obtain a representation vector that accurately reflects the characteristics of device behavior.

[0079] The above method uses high-quality behavioral representation vectors, enabling the model to calculate behavioral similarity more accurately, providing more precise data support for channel behavior quality assessment, and improving the ability to identify fake traffic.

[0080] In S514, the purpose of same-dimensional processing is to ensure consistency of device behavior data from different channels, thereby avoiding errors caused by data dimension mismatch, significantly improving the accuracy of subsequent device behavior similarity calculations, and ensuring the reliability of channel evaluation results.

[0081] Specifically, in some embodiments, the specific implementation of step S514 can be found in the following embodiments. This embodiment is based on... Figure 2 The detailed description of step S514 in the channel behavior quality assessment method shown in the corresponding embodiment includes the following steps:

[0082] Each device behavior chain corresponding to each channel is supplemented to obtain the supplemented device behavior chain.

[0083] Align the device behavior chains corresponding to each channel with the same position to obtain the same-dimensional device behavior chains.

[0084] In this embodiment, by using completion processing and same-position alignment processing, the problems of length differences and time misalignment between different channel and device behavior chains are effectively solved, ensuring the consistency of device behavior chains in length and position. This improves the accuracy of behavior similarity calculation and the precision of channel quality assessment, ultimately enabling more reliable identification of fake traffic and providing a more accurate basis for channel quality assessment.

[0085] Specifically, the completion process ensures the consistency of the length of the behavior chains across all devices, allowing data to be compared on the same dimension and eliminating interference caused by inconsistent behavior chain lengths. Simultaneously, the alignment process ensures that behavior data from different devices can be compared at the same point in time, making the comparison between device behavior chains more fair and accurate.

[0086] After completion and alignment, the device behavior chain has a uniform length and position, avoiding errors caused by data inconsistencies and thus improving the accuracy of behavior similarity calculation. Ultimately, the assessed channel traffic quality is more accurate, and the ability to identify fake traffic is stronger.

[0087] Whether it's the difference in behavioral data dimensions between channels or the misalignment of behavioral data length and time between devices, the application of completion and alignment technologies can handle a variety of complex situations, enabling the method to run stably in various scenarios.

[0088] In S516, by accurately calculating the similarity of device behavior, it is possible to determine whether the behavior of different devices within the same channel is consistent. If the similarity of behavior of most devices is high, it indicates that the traffic of that channel is relatively genuine and consistent; if the similarity is low, it may indicate the presence of fake traffic or low-quality traffic.

[0089] In S520, the calculation of cohesion and diversity values ​​helps to refine the evaluation of channel quality, rather than relying solely on simple metrics such as new users or ROI. This effectively reduces the impact of fraudulent traffic, making channel quality assessments more accurate. For example, if a channel has low cohesion and excessively high diversity, it may mean that the channel's traffic quality is low (e.g., there is a large amount of fraudulent or bot traffic). Conversely, channels with high cohesion and moderate diversity are more likely to represent high-quality, genuine user traffic.

[0090] Specifically, in some embodiments, the specific implementation of step S520 can be found in the following embodiments. This embodiment is based on... Figure 2 The detailed description of step S520 in the channel behavior quality assessment method shown in the corresponding embodiment includes the following steps:

[0091] The channel cohesion value is obtained by averaging the similarity of the behaviors of each device corresponding to each channel.

[0092] The channel diversity value is obtained by taking the standard deviation of the similarity of the device behaviors corresponding to each channel.

[0093] In this embodiment, cohesion reflects the degree of similarity in user behavior within the same channel. Specifically, a cohesion index is calculated by analyzing the similarity of device behavior within the channel. This index measures whether user behavior within the channel is consistent. A higher cohesion value indicates more uniform user behavior within the channel and higher traffic quality. In this embodiment, the cohesion value is obtained by taking the average value.

[0094] Suppose we have a channel A. We calculate the similarity values ​​among all its devices and then take the average of these similarity values ​​to obtain the cohesion value of that channel. This value clearly shows the degree of consistency in the behavior of the devices within that channel.

[0095] For example: Channel A has 5 devices with similarities of 0.8, 0.9, 0.85, 0.87, and 0.91. Calculate their mean, which is the cohesion value. :

[0096]

[0097] The diversity value reflects the differences in user behavior within the same channel. By calculating the differences in device behavior within a channel, the diversity index of that channel can be derived. A higher diversity value indicates that the user behavior of that channel covers different user needs, potentially meaning that the traffic sources of that channel are more diverse. In this embodiment, the cohesion value is obtained by taking the standard deviation.

[0098] Similarly, for channel A, the device behavior similarity values ​​are 0.8, 0.9, 0.85, 0.87, and 0.91. First, the mean is calculated to be 0.866. Then, the sum of squared deviations of each device behavior similarity from the mean is calculated, and finally, the standard deviation is 0.039. This is the diversity value of the channel.

[0099] This embodiment calculates the mean and standard deviation of device behavior similarity to obtain channel cohesion and channel diversity values, effectively solving the problem of how to quantify channel quality and evaluate device behavior characteristics. Channel cohesion and diversity values ​​allow for a better understanding of the performance characteristics of each channel, enabling optimization decisions and more efficient management and resource allocation.

[0100] In other embodiments, channel cohesion and channel diversity values ​​can also be obtained by classifying device behavior using clustering algorithms (such as K-means or DBSCAN), grouping similar device behaviors together, and then obtaining the results based on the clustering results.

[0101] Channel cohesion and channel diversity values ​​can also be obtained using methods such as spectral clustering based on similarity matrices, behavioral path analysis, behavioral pattern matching and hierarchical analysis, graph-based centrality analysis, behavioral frequency distribution, and information entropy.

[0102] In S530, the Channel Behavior Quality Score provides a comprehensive, quantitative assessment result that helps businesses select high-performing channels and avoid interference from low-quality or fraudulent traffic channels. The scoring mechanism helps businesses evaluate channels from multiple dimensions (such as traffic consistency, breadth, and diversity) and optimize advertising investment and traffic selection based on the score, thereby improving return on investment (ROI).

[0103] Specifically, in some embodiments, the specific implementation of step S530 can be found in [reference needed]. Figure 4 . Figure 4 It is based on Figure 2 According to the detailed description of step S530 in the channel behavior quality assessment method shown in the corresponding embodiment, step S530 in the channel behavior quality assessment method may include the following steps:

[0104] S532, Based on the channel cohesion benchmark value and channel cohesion value corresponding to each channel, determine the channel behavior cohesion score corresponding to each channel, wherein the channel cohesion benchmark value is the cohesion value determined from natural traffic.

[0105] S534, Based on the channel diversity benchmark value and channel diversity value corresponding to each channel, determine the channel behavior diversity score corresponding to each channel, wherein the channel diversity benchmark value is the diversity value determined from natural traffic.

[0106] S536. Based on the channel behavior cohesion score and the channel behavior diversity score corresponding to each channel, the channel behavior quality score corresponding to each channel is obtained.

[0107] In this embodiment, by introducing channel benchmark values, the cohesion and diversity scores of each channel are compared with benchmark values ​​determined in organic traffic. This ensures that the evaluation of channel behavior not only relies on the mean and standard deviation of internal data but also takes into account the external environment (i.e., performance in organic traffic). This approach effectively improves the reliability and relevance of channel evaluation, thereby providing a more accurate basis for subsequent strategy optimization.

[0108] In S532, the channel cohesion benchmark value is a cohesion value obtained from natural flow data, representing a standard or ideal channel behavior cohesion value.

[0109] Specifically, in some embodiments, the specific implementation of step S532 can be found in the following embodiments. This embodiment is based on... Figure 4 The detailed description of step S532 in the channel behavior quality assessment method shown in the corresponding embodiment includes the following steps:

[0110] Compare the channel cohesion benchmark value with the corresponding channel cohesion value for each channel.

[0111] If the channel cohesion value corresponding to the channel is smaller than the channel cohesion benchmark value, then the channel cohesion value is used as the difference in cohesion behavior magnitude.

[0112] If the channel cohesion value corresponding to the channel is greater than the channel cohesion benchmark value, then the difference between the unit value and the channel cohesion value is taken as the difference in cohesion behavior magnitude.

[0113] Determine the maximum and minimum cohesive behavior amplitude differences among the cohesive behavior amplitude differences corresponding to each channel.

[0114] Based on the maximum cohesive behavior amplitude difference, the minimum cohesive behavior amplitude difference, and the channel cohesion value corresponding to each channel, the channel behavior cohesion score is obtained.

[0115] In this embodiment, by introducing an amplitude difference mechanism, not only are numerical differences considered, but also situations where the cohesion value is too large or too small are addressed. This avoids the bias caused by simply relying on numerical comparisons and improves the accuracy of the evaluation. Simultaneously, by introducing a comparison between the maximum and minimum amplitude differences, different types of channels can be reasonably evaluated, and adjustments can be made appropriately regardless of whether their cohesion value is greater or less than the benchmark value. In other words, this embodiment, through dynamically calculating the amplitude difference, can adapt to the performance of different channels, effectively addressing different types of data and distributions that may be encountered in real-world environments, and providing more practical evaluation results.

[0116] Specifically, the channel behavior cohesion score can be obtained by the following formula:

[0117]

[0118]

[0119]

[0120] in, The channel cohesion value corresponding to each channel. As a benchmark value for channel cohesion, This indicates a channel cohesion value greater than the channel cohesion benchmark value. This indicates a channel cohesion value that is less than the channel cohesion benchmark value. This represents the difference in amplitude of cohesive behavior. In order to be in The maximum difference in cohesive behavior amplitude under the given conditions. In order to be in The minimum cohesive behavior amplitude difference under the given conditions. In order to be in The maximum difference in cohesive behavior amplitude under the given conditions. In order to be in The minimum cohesive behavior amplitude difference under the given conditions. The score represents the cohesion of channel behavior.

[0121] In the formula above, when the channel's cohesion value is less than the benchmark value, it may indicate poor consistency in the channel's equipment behavior. In this case, using the cohesion value itself as the magnitude difference can more accurately reflect the size of the gap. When the channel's cohesion value is greater than the benchmark value, it usually indicates strong consistency in the channel's equipment behavior. In this case, it is necessary not only to reflect the size of the gap but also to avoid overstating it. Therefore, the difference is calculated by subtracting the cohesion value from the unit value to balance this situation. Simultaneously, by distinguishing... and The calculation of the difference in cohesive behavior amplitude under different circumstances ensures that the evaluation of each channel not only considers its difference from the benchmark value, but also reasonably handles the situation where the cohesive value is too large or too small, thus reflecting the quality of channel behavior more comprehensively.

[0122] In some other embodiments, the channel behavior cohesion score can also be obtained directly from the ratio of the channel cohesion value to the channel cohesion benchmark value, or directly from the difference between the channel cohesion value and the channel cohesion benchmark value.

[0123] For example, in some embodiments, the channel behavior cohesion score can be obtained by the following formula:

[0124]

[0125] in, The channel cohesion value corresponding to each channel. As a benchmark value for channel cohesion, The score represents the cohesion of channel behavior.

[0126] In S534, the channel diversity benchmark value is still the standard value of channel behavior diversity obtained from natural traffic data, representing the ideal level of diversity.

[0127] Specifically, in some embodiments, the specific implementation of step S534 can be found in the following embodiments. This embodiment is based on... Figure 4 The detailed description of step S534 in the channel behavior quality assessment method shown in the corresponding embodiment includes the following steps:

[0128] Compare the benchmark value for channel diversity with the corresponding channel diversity values ​​for each channel.

[0129] If the channel diversity value corresponding to the channel is smaller than the channel diversity benchmark value, then the channel diversity value is used as the difference in diversity behavior magnitude.

[0130] If the channel diversity value corresponding to the channel is greater than the channel diversity benchmark value, then the difference between the unit value and the channel diversity value is used as the difference in diversity behavior magnitude.

[0131] Determine the maximum and minimum differences in the magnitude of diversity behaviors among the differences in the magnitude of diversity behaviors corresponding to each channel.

[0132] The channel behavior diversity score is obtained based on the maximum diversity behavior amplitude difference, the minimum diversity behavior amplitude difference, and the channel diversity value corresponding to each channel.

[0133] In this embodiment, by comparing the channel diversity value and the channel diversity benchmark value, the diversity differences among various channels can be quantified, avoiding overestimation or neglect of differences. Furthermore, by calculating the magnitude difference and determining the maximum and minimum differences, a diversity score is ultimately derived. This behavioral diversity score, calculated through these differences, accurately reflects the strengths and weaknesses of channels in terms of diversity, enabling precise and efficient evaluation of the quality of channel behavior. The behavioral diversity score is a comprehensive assessment score that helps optimize channel behavior management and decision-making.

[0134] Specifically, the channel behavior diversity score can be obtained by the following formula:

[0135]

[0136]

[0137]

[0138] in, This represents the channel diversity value for each channel. As a benchmark for channel diversity, This indicates a channel diversity value that is greater than the channel diversity benchmark value. This indicates a channel diversity value that is less than the channel diversity benchmark value. For the difference in the magnitude of diverse behaviors, In order to be in The maximum difference in the magnitude of diverse behaviors under the given conditions In order to be in The minimum difference in the magnitude of diverse behaviors under the given conditions. In order to be in The maximum difference in the magnitude of diverse behaviors under the given conditions In order to be in The minimum difference in the magnitude of diverse behaviors under the given conditions. Score the diversity of channel behavior.

[0139] In the above formula, if the channel's diversity level is lower than the benchmark value, the difference is directly measured using the channel's diversity value as a measure of its diversity behavior, which is beneficial for quickly assessing channels with low diversity. When the channel's diversity value is higher than the benchmark value, a single unit difference is used for calculation, avoiding excessive inflation of differences exceeding the benchmark and ensuring that the assessment results are reasonable and balanced. Simultaneously, by distinguishing... and The calculation of the difference in the range of diversity behaviors under different circumstances ensures that the evaluation of each channel not only takes into account its difference from the benchmark value, but also reasonably handles the situation where the diversity value is too large or too small, thus reflecting the quality of channel behavior more comprehensively.

[0140] In some other embodiments, the channel behavior diversity score can also be obtained directly from the ratio of the channel diversity value to the channel diversity benchmark value, or directly from the difference between the channel diversity value and the channel diversity benchmark value.

[0141] For example, in some embodiments, the channel behavior diversity score can be obtained by the following formula:

[0142]

[0143] in, This represents the channel diversity value for each channel. As a benchmark for channel diversity, Score the diversity of channel behavior.

[0144] In S536, a comprehensive channel behavior quality score is obtained by combining the cohesion score and the diversity score. The comprehensive channel behavior quality score can be calculated through weighted average, product, or other mathematical methods. Its purpose is to provide a more comprehensive and accurate quality score for each channel's performance by evaluating both cohesion and diversity, making the channel evaluation results more comparable and referential.

[0145] Specifically, in some embodiments, the specific implementation of step S536 can be found in the following embodiments. This embodiment is based on... Figure 4 The detailed description of step S536 in the channel behavior quality assessment method shown in the corresponding embodiment includes the following steps:

[0146] The channel behavior cohesion score and channel behavior diversity score for each channel are weighted and summed to obtain the channel behavior quality score for each channel.

[0147] In this embodiment, when assessing channel behavior quality, a single dimension (cohesion or diversity) may not fully reflect the overall performance of the channel. Cohesion primarily reflects the consistency or concentration of channel behavior, while diversity reflects the breadth or dispersion of channel behavior; the two have different focuses. By using a weighted summation method, the influence of both aspects can be comprehensively considered, making the assessment results more comprehensive and accurate. Furthermore, the impact of cohesion and diversity scores on quality assessment may differ across channels; using a weighted summation method allows for flexible adjustment of the weights of cohesion and diversity according to different business needs, thereby achieving personalized assessment.

[0148] Specifically, the channel behavior quality score can be obtained by the following formula:

[0149]

[0150] in, Assess the quality of channel behavior. For weighting percentage, For channel behavior cohesion score, Score the diversity of channel behavior.

[0151] In the above formula, the weight percentage can be a pre-set fixed value, such as 0.4, 0.5, 0.6, etc.

[0152] In other embodiments, the weighting percentage can also be dynamically determined based on the channel benchmark value, channel cohesion value, and channel diversity value.

[0153] For example, the weights can be obtained by normalizing the absolute values ​​of the difference between channel cohesion and channel benchmark values ​​and the absolute values ​​of the difference between channel diversity and channel benchmark values.

[0154] For example, the weights can be obtained by normalizing the ratio of channel cohesion value to channel benchmark value and the ratio of channel diversity value to channel benchmark value.

[0155] For example, channel benchmark values, channel cohesion values, and channel diversity values ​​can be input into the corresponding weight model, and the weight model can output the corresponding weight values.

[0156] Figure 5 A flowchart of a channel quality assessment method according to an embodiment of this application is shown. This channel quality assessment method can be executed by a server, which can also be... Figure 1 The server shown. (Refer to...) Figure 5 As shown, the channel quality assessment method includes at least the following:

[0157] S100, Obtain channel evaluation data, which includes channel new addition data, channel retention data, channel revenue data, and channel behavior data.

[0158] S200, Based on the newly added data of the channel, determine the quality score of the newly added channel.

[0159] S300, determine the channel retention quality score based on the channel retention data.

[0160] S400, determine the channel revenue quality score based on the channel revenue data.

[0161] S500, Based on the channel behavior data, determine the channel behavior quality score using the channel behavior quality assessment method described above.

[0162] S600, determine the channel quality assessment score based on the channel new user quality score, the channel retention quality score, the channel revenue quality score, and the channel behavior quality score.

[0163] In the embodiments of this application, by combining data from multiple dimensions such as new user acquisition, retention, revenue, and behavior, the problem that a single-dimensional assessment cannot comprehensively evaluate channel quality is solved, making channel quality assessment more comprehensive and accurate. Simultaneously, by integrating scores from different dimensions, the data from these dimensions is effectively combined and a unified quality score is formed through an appropriate weighting method, avoiding computational complexity and data redundancy, thus improving assessment efficiency and the practicality of the results. Furthermore, using a weighted summation method allows for flexible adjustment of the weights of each dimension according to different business needs, enabling the assessment results to better adapt to specific business scenarios.

[0164] In S100, the acquisition of channel evaluation data typically relies on extracting relevant data from multiple data sources such as channel operation system, user behavior analysis system, and financial system. This data may include user registration volume, number of active users, retention rate, revenue data, and specific channel behavior logs.

[0165] Among these metrics, channel acquisition data reflects the ability to attract new users and expand the channel. Channel retention data reflects user loyalty and long-term appeal to the channel. Channel revenue data measures the channel's economic efficiency or profitability.

[0166] In S200, the new user quality score may be based on factors such as the number of new users, user growth rate, and user origin, combined with certain standards or models to quantify the score. Typically, this data is compared with historical data to assess the stability and sustainability of new user acquisition capabilities.

[0167] Specifically, in some embodiments, the specific implementation of step S200 can be found in the following embodiments. This embodiment is based on... Figure 5 According to the detailed description of step S200 in the channel behavior quality assessment method shown in the corresponding embodiment, step S200 in the channel behavior quality assessment method may include the following steps:

[0168] Based on the newly added data from the aforementioned channels, the scores for newly established channels and existing channels are determined separately.

[0169] The new channel quality score is determined based on the new score of the newly established channel and the new score of the existing channel.

[0170] In this embodiment, by distinguishing between the new scores of newly established channels and existing channels, and calculating the new scores of newly established channels and existing channels separately, the new quality of different types of channels can be evaluated more accurately, avoiding errors caused by uniform evaluation of channel quality. It can also reflect the actual performance of different channel types in terms of new data in more detail, thereby improving the scientificity, accuracy and operability of the evaluation results.

[0171] Newly added channels generally refer to those that have recently joined. The quality assessment of their newly added data may be influenced by various factors, such as initial user activity, data growth rate, and data completeness. Based on these factors, a corresponding new channel score is obtained to reflect the quality of the newly added data. For example, key indicators (such as data growth rate and number of active users) can be set to measure the quality of the new data, thereby deriving the new channel score.

[0172] Existing channels typically refer to channels that have been in existence for some time, and the quality of their new data may differ from that of newly established channels. The new data score for existing channels may be affected by the channel's long-term operational performance; therefore, it is necessary to evaluate the quality of their new data by comparing historical data, new data, and existing data.

[0173] For example, you can compare the new data of a channel with its past performance, the growth rate of the data, user feedback and other factors to obtain the new data score for an existing channel.

[0174] The final channel addition quality score is determined by combining the scores of newly established channels and existing channels. This can be achieved through methods such as weighted average, ranking, and segmented scoring, with the specific calculation method depending on the specific implementation plan.

[0175] In S300, retention quality score typically relies on metrics such as user activity, usage frequency, and retention rate. Based on the continued retention of users over a specific time period, the channel's retention quality score can be calculated. The model may use different time periods (e.g., 1 week, 1 month, 1 year) to calculate the retention rate.

[0176] Specifically, in some embodiments, the specific implementation of step S300 can be found in the following embodiments. This embodiment is based on... Figure 5 According to the detailed description of step S300 in the channel behavior quality assessment method shown in the corresponding embodiment, step S300 in the channel behavior quality assessment method may include the following steps:

[0177] Based on the retention data from the aforementioned channels, determine the next-day retention rate, weekly retention rate, and long-term retention rate.

[0178] The channel retention quality score is determined based on the next-day retention rate, the weekly retention rate, and the long-term retention rate.

[0179] In this embodiment, retention data helps channel operators identify time points or stages with low retention rates, enabling targeted optimization measures to improve overall channel performance. By analyzing the channel's day-one retention rate, weekly retention rate, and long-term retention rate, user retention can be measured from multiple perspectives. This multi-dimensional evaluation method accurately reflects user stickiness and long-term value at different time stages, avoiding the bias that may arise from single-dimensional indicators. Retention rates at different time dimensions reflect different levels of channel performance. For example, day-one retention rate focuses on short-term user stickiness, weekly retention rate reflects the channel's medium-term user retention ability, and long-term retention rate indicates the channel's long-term operational effectiveness. By integrating data from these dimensions, a more comprehensive and reliable channel retention quality score can be obtained.

[0180] Day 2 retention rate typically refers to the percentage of new users who continue using the channel the day after installation or initial exposure. It reflects users' initial interest and short-term engagement. A high day 2 retention rate indicates strong short-term appeal to the channel. Weekly retention rate represents the percentage of users who continue using the channel within a specific week after their initial use. Weekly retention rate better reflects users' medium-term engagement. It considers not only whether users maintain their interest in the channel but also assesses the long-term appeal of the channel's content or service. Long-term retention rate typically refers to the percentage of users who remain active after a considerable period (e.g., several months) following their initial exposure to the channel. This metric reflects the long-term value of the channel, and a high long-term retention rate usually indicates good performance in continuous optimization and user experience.

[0181] Specifically, each retention rate can be standardized to eliminate scaling differences between data, allowing for fair comparisons of retention rates across different time dimensions. Then, depending on business needs, different weights may be assigned to different retention rate dimensions. For example, short-term retention rates (such as day-one retention rates) may be more important for some industries, while long-term retention rates better reflect the overall stability of a channel. Therefore, different weighting coefficients can be set based on channel characteristics and objectives. Finally, by combining the weighted retention rate data, methods such as weighted averaging, segmented scoring, and ranking can be used to comprehensively calculate the channel's final retention quality score. The resulting channel retention quality score, as part of the channel behavior quality assessment, provides quantitative feedback on its long-term performance and potential.

[0182] In S400, revenue quality scores are typically calculated based on the channel's direct economic benefits, such as sales volume, profit, and cost-effectiveness. Revenue data may come from sales platforms, financial systems, etc., and the score will consider factors such as revenue growth, stability, and cost-effectiveness.

[0183] Specifically, in some embodiments, the specific implementation of step S400 can be found in the following embodiments. This embodiment is based on... Figure 5 According to the detailed description of step S400 in the channel behavior quality assessment method shown in the corresponding embodiment, step S400 in the channel behavior quality assessment method may include the following steps:

[0184] The active device revenue, channel revenue, and channel cost are determined based on the channel revenue data.

[0185] The channel revenue quality score is determined based on the active device revenue, the channel revenue, and the channel cost.

[0186] In this embodiment, by considering multiple dimensions such as active device revenue, channel revenue, and channel costs, and reasonably combining various data, a more comprehensive revenue quality assessment can be provided, reducing bias. Simultaneously, it achieves precise quantification of channel revenue quality, thus providing decision-makers with more scientific and actionable evaluation results. This embodiment can identify which channels have high revenue quality and which channels may have problems such as poor cost control or unstable revenue, thereby helping operators optimize channel structure and resource allocation.

[0187] Specifically, active device revenue refers to the revenue contributed by active devices (such as user terminal devices) associated with a channel, typically considering factors such as device activity and usage, including online time and frequency of use. Channel revenue is the total revenue generated by the channel through all devices, usually including revenue generated through sales, advertising, or other business models. Channel costs are all costs incurred by the channel for its operation, including equipment investment, maintenance fees, promotional expenses, and other data that reveal the channel's operational efficiency and cost control.

[0188] For details regarding the S500, please refer to [link / reference]. Figures 2 to 4 The channel behavior quality assessment methods mentioned will not be elaborated upon here.

[0189] In S600, a weighted summation method is typically used, weighting scores for new user acquisition, retention, revenue, and behavior according to preset weights to arrive at an overall channel quality score. The weighting coefficients can be adjusted based on business needs to highlight the importance of certain dimensions. For example, for rapidly expanding channels, new user acquisition scores may carry more weight; for long-term operating channels, retention and revenue scores may be more crucial.

[0190] Specifically, in some embodiments, the specific implementation of step S600 can be found in the following embodiments. This embodiment is based on... Figure 5 According to the detailed description of step S600 in the channel behavior quality assessment method shown in the corresponding embodiment, step S600 in the channel behavior quality assessment method may include the following steps:

[0191] The channel quality assessment score is obtained by weighting and summing the channel new user quality score, channel retention quality score, channel revenue quality score, and channel behavior quality score.

[0192] In this embodiment, by weighting and summing the quality scores across different dimensions (new user quality, retention quality, revenue quality, and behavioral quality), the overall performance of the channel can be evaluated more comprehensively and accurately. This method effectively addresses the limitations of single-indicator evaluation and, through flexible weight allocation, ensures that the final evaluation results better align with actual operational needs, thereby providing decision-makers with more scientific support and helping to optimize channel management and resource allocation.

[0193] In some embodiments, the weighting coefficients for each dimension can be pre-set fixed empirical values, such as 0.25, 0.2, 0.25, and 0.3.

[0194] In other embodiments, the weighting coefficients for each dimension can also be dynamically determined based on the channel benchmark value, channel cohesion value, and channel diversity value.

[0195] For example, the weight coefficients of each dimension can be determined based on the mean and dispersion of the quality scores of each dimension. The weight coefficient of a dimension is positively correlated with its dispersion.

[0196] The following describes an embodiment of the apparatus described in this application, which can be used to execute the channel behavior quality assessment method and channel quality assessment method described in the above embodiments of this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the channel behavior quality assessment method and channel quality assessment method described above in this application.

[0197] Figure 6 A block diagram of a channel behavior quality assessment apparatus according to an embodiment of this application is shown.

[0198] Reference Figure 6 As shown, a channel behavior quality assessment device 150 according to an embodiment of this application includes: a device similarity determination module 151, a cohesion diversity determination module 152, and a behavior score determination module 153.

[0199] The device similarity determination module 151 is used to obtain the device behavior similarity between multiple devices corresponding to each channel based on the channel behavior data of each channel; the cohesion and diversity determination module 152 is used to determine the channel cohesion value and channel diversity value corresponding to each channel based on the device behavior similarity corresponding to each channel; and the behavior score determination module 153 is used to determine the channel behavior quality score corresponding to each channel based on the channel benchmark value, channel cohesion value and channel diversity value corresponding to each channel.

[0200] In some embodiments of this application, the device similarity determination module specifically includes: a behavior representation extraction submodule, used to extract device behavior representations from the channel behavior data of each channel using a behavior representation extraction model to obtain device behavior chains; a device same-dimensional processing submodule, used to perform same-dimensional processing on each device behavior chain corresponding to each channel to obtain corresponding same-dimensional device behavior chains; and a device similarity determination submodule, used to compare each same-dimensional device behavior chain corresponding to each channel to obtain the device behavior similarity between devices.

[0201] In some embodiments of this application, the device same-dimensional processing submodule specifically includes: a behavior chain completion unit, used to complete each device behavior chain corresponding to each channel to obtain a completed device behavior chain; and a same-position alignment unit, used to perform same-position alignment processing on each device behavior chain corresponding to each channel to obtain a same-dimensional device behavior chain.

[0202] In some embodiments of this application, the behavior representation extraction model includes an encoder and a decoder, and the channel behavior quality assessment device further includes: a sample acquisition module for acquiring a set of device behavior data sequence samples, wherein the set of device behavior data sequence samples contains multiple device behavior data sequence samples, and each device behavior data sequence sample is labeled with a corresponding behavior representation sequence; a sample encoding module for inputting the device behavior data sequence samples into the encoder to obtain behavior sequence representations; a sample decoding module for inputting the behavior sequence representations into the decoder to obtain behavior representation vectors; a loss function module for determining a loss function based on the finally obtained behavior representation vectors and the labeled behavior representation sequences; and a parameter update module for updating the parameters of the encoder and the decoder based on the loss function until the loss function converges.

[0203] In some embodiments of this application, the cohesion diversity determination module specifically includes: a channel cohesion value submodule, used to take the mean of the similarity of the behavior of each device corresponding to each channel to obtain a channel cohesion value; and a channel diversity value submodule, used to take the standard deviation of the similarity of the behavior of each device corresponding to each channel to obtain a channel diversity value.

[0204] In some embodiments of this application, the channel benchmark value includes a channel cohesion benchmark value and a channel diversity benchmark value. The behavior score determination module specifically includes: a behavior cohesion numerator module, used to determine the channel behavior cohesion score corresponding to each channel based on the channel cohesion benchmark value and the channel cohesion value corresponding to each channel, wherein the channel cohesion benchmark value is a cohesion value determined from natural traffic; a behavior diversity numerator module, used to determine the channel behavior diversity score corresponding to each channel based on the channel diversity benchmark value and the channel diversity value corresponding to each channel, wherein the channel diversity benchmark value is a diversity value determined from natural traffic; and a channel behavior quality numerator module, used to obtain the channel behavior quality score corresponding to each channel based on the channel behavior cohesion score and the channel behavior diversity score corresponding to each channel.

[0205] In some embodiments of this application, the behavioral cohesion module specifically includes: a cohesion benchmark comparison unit, used to compare the channel cohesion benchmark value with the channel cohesion value corresponding to each channel; a first cohesion amplitude unit, used to use the channel cohesion value as the cohesion behavioral amplitude difference if the channel cohesion value corresponding to the channel is smaller than the channel cohesion benchmark value; a second cohesion amplitude unit, used to use the difference between the unit value and the channel cohesion value as the cohesion behavioral amplitude difference if the channel cohesion value corresponding to the channel is larger than the channel cohesion benchmark value; a cohesion amplitude extreme value unit, used to determine the maximum and minimum cohesion behavioral amplitude differences among the cohesion behavioral amplitude differences corresponding to each channel; and a cohesion score calculation unit, used to obtain the channel behavioral cohesion score based on the maximum cohesion behavioral amplitude difference, the minimum cohesion behavioral amplitude difference, and the channel cohesion value corresponding to each channel.

[0206] In some embodiments of this application, the behavioral diversity module specifically includes: a diversity benchmark comparison unit, used to compare the channel diversity benchmark value with the channel diversity value corresponding to each channel; a first diversity amplitude unit, used to use the channel diversity value as the diversity behavioral amplitude difference if the channel diversity value corresponding to the channel is smaller than the channel diversity benchmark value; a second diversity amplitude unit, used to use the difference between the unit value and the channel diversity value as the diversity behavioral amplitude difference if the channel diversity value corresponding to the channel is larger than the channel diversity benchmark value; a diversity amplitude extreme value unit, used to determine the maximum and minimum diversity behavioral amplitude differences among the diversity behavioral amplitude differences corresponding to each channel; and a diversity score calculation unit, used to obtain a channel behavioral diversity score based on the maximum diversity behavioral amplitude difference, the minimum diversity behavioral amplitude difference, and the channel diversity value corresponding to each channel.

[0207] In some embodiments of this application, the channel behavior quality module specifically includes: a channel behavior quality scoring unit, used to perform a weighted summation of the channel behavior cohesion score and the channel behavior diversity score corresponding to each channel to obtain the channel behavior quality score corresponding to each channel.

[0208] In this embodiment, by analyzing the similarity of device behavior, the consistency of traffic within a channel is assessed, helping to identify fraudulent traffic. Simultaneously, by combining behavioral consistency and behavioral differences with benchmark comparisons, a comprehensive quality score is derived for each channel. This allows for a more comprehensive analysis of channel traffic quality, aiding in optimization decisions and effectively addressing the technical limitations of traditional evaluation methods. By introducing behavioral data and multi-dimensional evaluation indicators, the quality evaluation system for user acquisition channels is enriched, maximizing the accuracy of the evaluation, optimizing the channel selection process, and ultimately improving the scientific rigor and efficiency of advertising placement and promotion.

[0209] Figure 7 A block diagram of a channel quality assessment system according to an embodiment of this application is shown.

[0210] Reference Figure 7 As shown, a channel quality assessment system 100 according to an embodiment of this application includes: a channel quality data acquisition device 110, a channel new addition quality assessment device 120, a channel retention quality assessment device 130, a channel revenue quality assessment device 140, a channel behavior quality assessment device 150, and a channel quality assessment scoring device 160.

[0211] The channel quality data acquisition device 110 is used to acquire channel evaluation data, including channel new addition data, channel retention data, channel revenue data, and channel behavior data. The channel new addition quality evaluation device 120 is used to determine a channel new addition quality score based on the channel new addition data. The channel retention quality evaluation device 130 is used to determine a channel retention quality score based on the channel retention data. The channel revenue quality evaluation device 140 is used to determine a channel revenue quality score based on the channel revenue data. The channel behavior quality evaluation device 150 is used to determine a channel behavior quality score based on the channel behavior data using the channel behavior quality evaluation method described above. Figure 6 The channel behavior quality assessment device 150 shown; the channel quality assessment scoring device 160 is used to determine the channel quality assessment score based on the channel new addition quality score, the channel retention quality score, the channel revenue quality score, and the channel behavior quality score.

[0212] In some embodiments of this application, the channel quality data acquisition device specifically includes: a channel addition score module, used to determine the new channel addition score and the existing channel addition score respectively based on the new channel data; and an addition quality score module, used to determine the new channel quality score based on the new channel addition score and the existing channel addition score.

[0213] In some embodiments of this application, the channel retention quality assessment device specifically includes: a retention parameter determination module, used to determine the next-day retention rate, weekly retention rate, and long-term retention rate based on the channel retention data; and a retention quality score module, used to determine the channel retention quality score based on the next-day retention rate, the weekly retention rate, and the long-term retention rate.

[0214] In some embodiments of this application, the channel revenue quality assessment device specifically includes: a revenue cost determination module, used to determine active device revenue, channel revenue, and channel cost based on the channel revenue data; and a revenue quality scoring module, used to determine a channel revenue quality score based on the active device revenue, the channel revenue, and the channel cost.

[0215] In some embodiments of this application, the channel quality assessment scoring device specifically includes: a channel quality assessment scoring module, used to perform a weighted summation of the channel new user quality score, the channel retention quality score, the channel revenue quality score, and the channel behavior quality score to obtain a channel quality assessment score.

[0216] In the embodiments of this application, by combining data from multiple dimensions such as new user acquisition, retention, revenue, and behavior, the problem that a single-dimensional assessment cannot comprehensively evaluate channel quality is solved, making channel quality assessment more comprehensive and accurate. Simultaneously, by integrating scores from different dimensions, the data from these dimensions is effectively combined and a unified quality score is formed through an appropriate weighting method, avoiding computational complexity and data redundancy, thus improving assessment efficiency and the practicality of the results. Furthermore, using a weighted summation method allows for flexible adjustment of the weights of each dimension according to different business needs, enabling the assessment results to better adapt to specific business scenarios.

[0217] Figure 8 A schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application is shown.

[0218] It should be noted that, Figure 8 The computer system of the electronic 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.

[0219] like Figure 8As shown, the computer system includes a Central Processing Unit (CPU) 1801, which can perform various appropriate actions and processes based on programs stored in Read-Only Memory (ROM) 1802 or programs loaded from storage portion 1808 into Random Access Memory (RAM) 1803, such as performing the methods described in the above embodiments. Various programs and data required for system operation are also stored in RAM 1803. The CPU 1801, ROM 1802, and RAM 1803 are interconnected via bus 1804. An Input / Output (I / O) interface 1805 is also connected to bus 1804.

[0220] The following components are connected to I / O interface 1805: an input section 1806 including a keyboard, mouse, etc.; an output section 1807 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 1808 including a hard disk, etc.; and a communication section 1809 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 1809 performs communication processing via a network such as the Internet. A drive 1810 is also connected to I / O interface 1805 as needed. Removable media 1811, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 1810 as needed so that computer programs read from them can be installed into storage section 1808 as needed.

[0221] 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 including a computer program 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 1809, and / or installed from removable medium 1811. When the computer program is executed by central processing unit (CPU) 1801, it performs various functions defined in the system of this application.

[0222] 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. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, 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 device, magnetic storage device, 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 a computer-readable computer program. The transmitted data signal 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 computer program 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.

[0223] 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.

[0224] 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.

[0225] In another aspect, this application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into the electronic device. The computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the methods described in the above embodiments.

[0226] This specification also provides a computer program product that stores at least one instruction, said at least one instruction being loaded and executed by the processor as described above. Figures 1-5 The method described in the illustrated embodiment can be found in the following document for a detailed execution process. Figures 1-5 The specific details of the illustrated embodiments will not be elaborated here.

[0227] 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.

[0228] 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 methods according to the embodiments of this application.

[0229] 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.

[0230] 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 method for assessing the quality of channel behavior, characterized in that, The channel behavior quality assessment method includes: Based on the channel behavior data for each channel, the device behavior similarity among multiple devices corresponding to the channel is obtained; Based on the similarity of the device behaviors corresponding to each channel, determine the channel cohesion value and channel diversity value corresponding to each channel; Based on the channel benchmark value, channel cohesion value, and channel diversity value corresponding to each channel, determine the channel behavior quality score corresponding to each channel; The step of obtaining the device behavior similarity among multiple devices corresponding to each channel based on the channel behavior data of each channel specifically includes: Device behavior representations are extracted from the channel behavior data of each channel using a behavior representation extraction model to obtain the device behavior chain; Perform same-dimensional processing on each device behavior chain corresponding to each channel to obtain the corresponding same-dimensional device behavior chain; The behavior chains of each device in the same dimension corresponding to each channel are compared to obtain the device behavior similarity between each device. If the device behavior similarity between each device corresponding to the channel is higher, the traffic of the channel is more real and consistent. The step of performing same-dimensional processing on each device behavior chain corresponding to each channel to obtain the corresponding same-dimensional device behavior chain specifically includes: Each device behavior chain corresponding to each channel is completed to obtain a completed device behavior chain, so as to ensure the consistency of the length of the behavior chains of all devices. Each device behavior chain corresponding to each channel is aligned to the same position to obtain a device behavior chain of the same dimension, so as to ensure that behavior data from different devices can be compared at the same time point; The step of determining the channel cohesion value and channel diversity value for each channel based on the similarity of device behavior corresponding to each channel specifically includes: The channel cohesion value is obtained by averaging the similarity of the device behaviors corresponding to each channel. The standard deviation of the similarity of the device behaviors corresponding to each channel is taken to obtain the channel diversity value; The channel benchmark values ​​include channel cohesion benchmark values ​​and channel diversity benchmark values. The determination of the channel behavior quality score for each channel based on its corresponding channel benchmark values, channel cohesion values, and channel diversity values ​​specifically includes: Based on the channel cohesion benchmark value and channel cohesion value corresponding to each channel, the channel behavior cohesion score corresponding to each channel is determined. The channel cohesion benchmark value is the cohesion value determined from natural traffic. Based on the channel diversity benchmark value and channel diversity value corresponding to each channel, the channel behavior diversity score corresponding to each channel is determined. The channel diversity benchmark value is the diversity value determined from natural traffic. Based on the channel behavior cohesion score and the channel behavior diversity score for each channel, the channel behavior quality score for each channel is obtained.

2. The channel behavior quality assessment method as described in claim 1, characterized in that, The behavior representation extraction model includes an encoder and a decoder, and the method further includes: Obtain a set of device behavior data sequence samples, wherein the set of device behavior data sequence samples contains multiple device behavior data sequence samples, and each device behavior data sequence sample is labeled with a corresponding behavior representation sequence; The device behavior data sequence sample is input into the encoder to obtain the behavior sequence representation; The behavior sequence representation is input into the decoder to obtain the behavior representation vector; Based on the final obtained behavior representation vector and the labeled behavior representation sequence, the loss function is determined; The encoder and decoder are updated based on the loss function until the loss function converges.

3. The channel behavior quality assessment method as described in claim 1, characterized in that, The process of determining the channel behavior cohesion score for each channel based on the channel cohesion benchmark value and channel cohesion value for each channel specifically includes: Compare the channel cohesion benchmark value with the corresponding channel cohesion value for each channel; If the channel cohesion value corresponding to the channel is smaller than the channel cohesion benchmark value, then the channel cohesion value is used as the difference in cohesion behavior magnitude. If the channel cohesion value corresponding to the channel is greater than the channel cohesion benchmark value, then the difference between the unit value and the channel cohesion value shall be used as the difference in cohesion behavior magnitude. Determine the maximum and minimum cohesive behavior amplitude differences among the cohesive behavior amplitude differences corresponding to each channel; Based on the maximum cohesive behavior amplitude difference, the minimum cohesive behavior amplitude difference, and the channel cohesion value corresponding to each channel, the channel behavior cohesion score is obtained.

4. The channel behavior quality assessment method as described in claim 1, characterized in that, The process of determining the channel behavior diversity score for each channel based on the channel diversity benchmark value and channel diversity value corresponding to each channel specifically includes: Compare the benchmark value for channel diversity with the corresponding channel diversity values ​​for each channel; If the channel diversity value corresponding to the channel is smaller than the channel diversity benchmark value, then the channel diversity value is used as the difference in diversity behavior magnitude. If the channel diversity value corresponding to the channel is greater than the channel diversity benchmark value, then the difference between the unit value and the channel diversity value shall be used as the difference in diversity behavior magnitude. Determine the maximum and minimum differences in the magnitude of diverse behaviors among the differences in the magnitude of diverse behaviors corresponding to each channel; The channel behavior diversity score is obtained based on the maximum diversity behavior amplitude difference, the minimum diversity behavior amplitude difference, and the channel diversity value corresponding to each channel.

5. The channel behavior quality assessment method as described in claim 1, characterized in that, The method of obtaining the channel behavior quality score for each channel based on the channel behavior cohesion score and channel behavior diversity score for each channel specifically includes: The channel behavior cohesion score and channel behavior diversity score for each channel are weighted and summed to obtain the channel behavior quality score for each channel.

6. A method for assessing channel quality, characterized in that, The channel quality assessment method includes: Obtain channel evaluation data, which includes channel new user data, channel retention data, channel revenue data, and channel behavior data; Based on the newly added data from the aforementioned channels, determine the quality score for newly added channels; Based on the channel retention data, a channel retention quality score is determined; Based on the channel revenue data, determine the channel revenue quality score; Based on the channel behavior data, the channel behavior quality score is determined using the channel behavior quality assessment method as described in any one of claims 1 to 5. The channel quality assessment score is determined based on the channel new user quality score, the channel retention quality score, the channel revenue quality score, and the channel behavior quality score.

7. The channel quality assessment method as described in claim 6, characterized in that, The step of determining the channel new data quality score specifically includes: Based on the newly added data of the channels, the new scores for newly established channels and the new scores for existing channels are determined respectively; The new channel quality score is determined based on the new score of the newly established channel and the new score of the existing channel.

8. The channel quality assessment method as described in claim 6, characterized in that, Based on the channel retention data, a channel retention quality score is determined, specifically including: Based on the retention data from the aforementioned channels, determine the next-day retention rate, weekly retention rate, and long-term retention rate; The channel retention quality score is determined based on the next-day retention rate, the weekly retention rate, and the long-term retention rate.

9. The channel quality assessment method as described in claim 6, characterized in that, The determination of the channel revenue quality score based on the channel revenue data specifically includes: The active device revenue, channel revenue, and channel cost are determined based on the channel revenue data. The channel revenue quality score is determined based on the active device revenue, the channel revenue, and the channel cost.

10. The channel quality assessment method as described in claim 6, characterized in that, The process of determining a channel quality assessment score based on the channel new user quality score, the channel retention quality score, the channel revenue quality score, and the channel behavior quality score specifically includes: The channel quality assessment score is obtained by weighting and summing the channel new user quality score, channel retention quality score, channel revenue quality score, and channel behavior quality score.

11. A channel behavior quality assessment device, characterized in that, The channel behavior quality assessment device includes: The device similarity determination module is used to obtain the device behavior similarity between multiple devices corresponding to each channel based on the channel behavior data of each channel. The cohesion and diversity determination module is used to determine the channel cohesion value and channel diversity value for each channel based on the similarity of the behaviors of each device corresponding to each channel. The behavior score determination module is used to determine the channel behavior quality score for each channel based on the channel benchmark value, channel cohesion value, and channel diversity value corresponding to each channel. The device similarity determination module specifically includes: a behavior representation extraction submodule, used to extract device behavior representations from the channel behavior data of each channel using a behavior representation extraction model to obtain device behavior chains; a device same-dimensional processing submodule, used to perform same-dimensional processing on each device behavior chain corresponding to each channel to obtain corresponding same-dimensional device behavior chains; and a device similarity determination submodule, used to compare each same-dimensional device behavior chain corresponding to each channel to obtain the device behavior similarity between devices. If the device behavior similarity between devices corresponding to a channel is higher, the traffic of that channel is more realistic and consistent. The device same-dimensional processing submodule specifically includes: a behavior chain completion unit, used to complete the behavior chain of each device corresponding to each channel to obtain a completed device behavior chain, so as to ensure the consistency of the length of the behavior chain of all devices; and a same-position alignment unit, used to perform same-position alignment processing on the behavior chain of each device corresponding to each channel to obtain a same-dimensional device behavior chain, so as to ensure that the behavior data from different devices can be compared at the same time node. The cohesion-diversity determination module specifically includes: a channel cohesion value submodule, used to take the mean of the similarity of the behavior of each device corresponding to each channel to obtain the channel cohesion value; and a channel diversity value submodule, used to take the standard deviation of the similarity of the behavior of each device corresponding to each channel to obtain the channel diversity value. The channel benchmark values ​​include channel cohesion benchmark values ​​and channel diversity benchmark values. The behavior score determination module specifically includes: a behavior cohesion sub-module, used to determine the channel behavior cohesion score for each channel based on the channel cohesion benchmark value and the channel cohesion value corresponding to each channel, wherein the channel cohesion benchmark value is the cohesion value determined from natural traffic; a behavior diversity sub-module, used to determine the channel behavior diversity score for each channel based on the channel diversity benchmark value and the channel diversity value corresponding to each channel, wherein the channel diversity benchmark value is the diversity value determined from natural traffic; and a channel behavior quality sub-module, used to obtain the channel behavior quality score for each channel based on the channel behavior cohesion score and the channel behavior diversity score corresponding to each channel.

12. A channel quality assessment system, characterized in that, The channel quality assessment system includes: A channel quality data acquisition device is used to acquire channel evaluation data, which includes channel new data, channel retention data, channel revenue data, and channel behavior data. A new channel quality assessment device is added to determine the new channel quality score based on the new channel data. A channel retention quality assessment device is used to determine a channel retention quality score based on the channel retention data. A channel revenue quality assessment device is used to determine a channel revenue quality score based on the channel revenue data. A channel behavior quality assessment device is used to determine a channel behavior quality score based on the channel behavior data using the channel behavior quality assessment method as described in any one of claims 1 to 5. The channel quality assessment scoring device is used to determine the channel quality assessment score based on the channel new addition quality score, the channel retention quality score, the channel revenue quality score, and the channel behavior quality score.

13. A computer-readable medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the channel behavior quality assessment method as described in any one of claims 1 to 5 or the channel quality assessment method as described in claims 6 to 10.

14. An electronic device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the channel behavior quality assessment method as described in any one of claims 1 to 5 or to implement the channel quality assessment method as described in claims 6 to 10.

15. A computer program product comprising one or more computer programs, characterized in that, When the one or more computer programs are executed by one or more processors, they implement the channel behavior quality assessment method as described in any one of claims 1 to 5 or the channel quality assessment method as described in any one of claims 6 to 10.

Images