User experience threshold determination method, content distribution method and related device

Abstract

The invention provides a user experience threshold determination method, a content distribution method and a related device, and relates to the field of artificial intelligence, in particular to the technical fields of content distribution, Internet, data processing and the like. The method for determining the user experience threshold comprises the following steps: acquiring content distribution posterior data and content distribution priori data corresponding to a target content provider; determining target user experience data corresponding to the target content provider according to the content distribution posterior data and the content distribution prior data; according to the target user experience data and the user experience mean value, a user experience threshold value corresponding to the target content provider is determined, the user experience threshold value is used for controlling distribution of content corresponding to the target content provider, and the user experience mean value is obtained based on the user experience data corresponding to the content provider. The content distribution effect can be effectively improved.

Description

Technical Field

[0001] The present disclosure relates to technical fields such as content distribution, the Internet, and data processing in the field of artificial intelligence, and in particular to a method for determining a user experience threshold, a content distribution method, and related devices. Background Art

[0002] With the continuous development of Internet multimedia, multimedia has become one of the important means of promoting products and services. Multimedia is, for example, soft advertising. Soft advertising requires the natural integration of advertising information with native content to reduce the audience's resistance. Therefore, there are relatively strict constraints on the user experience of soft advertising, such as strict requirements on indicators such as click-through rate and active user time.

[0003] At present, the multimedia content distribution effect is usually improved by adjusting the base price of the unit multimedia traffic or adjusting the user experience constraints. However, the above methods have the problem that the improvement of the content distribution effect is not obvious. Summary of the invention

[0004] The present disclosure provides a method for determining a user experience threshold for improving multimedia content distribution effects, a content distribution method, and related devices.

[0005] According to a first aspect of the present disclosure, a method for determining a user experience threshold is provided, including: obtaining content distribution posterior data and content distribution prior data corresponding to a target content provider; determining target user experience data corresponding to the target content provider based on the content distribution posterior data and the content distribution prior data; determining a user experience threshold corresponding to the target content provider based on the target user experience data and a user experience mean, the user experience threshold being used to control the distribution of content corresponding to the target content provider, and the user experience mean being obtained based on the user experience data corresponding to the content provider.

[0006] According to a second aspect of the present disclosure, a content distribution method is provided, comprising: obtaining, from a material library, content to be distributed corresponding to a content provider according to a user experience threshold corresponding to the content provider, wherein the user experience threshold is determined using the method for determining the user experience threshold as described in the first aspect of the present disclosure; and distributing the content to be distributed.

[0007] According to a third aspect of the present disclosure, a device for determining a user experience threshold is provided, including: an acquisition unit, used to acquire content distribution posteriori data and content distribution priori data corresponding to a target content provider; a first determination unit, used to determine target user experience data corresponding to the target content provider based on the content distribution posteriori data and the content distribution priori data; a second determination unit, used to determine the user experience threshold corresponding to the target content provider based on the target user experience data and the user experience mean, the user experience threshold being used to control the distribution of content corresponding to the target content provider, and the user experience mean being obtained based on the user experience data corresponding to the content provider.

[0008] According to a fourth aspect of the present disclosure, a content distribution device is provided, comprising: an acquisition unit, used to acquire, from a material library, content to be distributed corresponding to a content provider according to a user experience threshold corresponding to the content provider, wherein the user experience threshold is determined using the method for determining the user experience threshold as described in the first aspect of the present disclosure; and a distribution unit, used to distribute the content to be distributed.

[0009] According to a fifth aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for determining the user experience threshold described in the first aspect or execute the content distribution method described in the second aspect.

[0010] According to a sixth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method for determining the user experience threshold described in the first aspect or execute the content distribution method described in the second aspect.

[0011] According to a seventh aspect of the present disclosure, a computer program product is provided, comprising: a computer program, the computer program being stored in a readable storage medium, at least one processor of an electronic device being able to read the computer program from the readable storage medium, the at least one processor executing the computer program so that the electronic device executes the method for determining the user experience threshold described in the first aspect or executes the content distribution method described in the second aspect.

[0012] The technology disclosed in the present invention solves the problem that the improvement of multimedia content distribution effect is not obvious when improving the content distribution effect through the current method. The present invention determines the target user experience data corresponding to the target content provider according to the content distribution posterior data and content distribution prior data corresponding to the target content provider, realizes the dynamic adjustment of the target user experience data, and is used to accurately determine the user experience threshold; compares the target user experience data with the user experience mean, thereby determining the user experience threshold corresponding to the target content provider, and realizes the dynamic adjustment of the user experience threshold. The user experience threshold is used to control the distribution of the content corresponding to the target content provider, which can optimize the traffic distribution of content distribution, thereby optimizing the balance between content delivery and user experience, and effectively improving the content distribution effect; and based on the target user experience data corresponding to the target content provider, the target content provider can be guided to provide higher quality content, thereby improving the overall content quality, and then the content distribution effect can be improved, achieving a win-win situation for the content distribution platform and the target content provider.

[0013] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0015] Figure 1 is a schematic diagram of an application scenario to which the present disclosure is applicable;

[0016] Figure 2 is a schematic diagram according to a first embodiment of the present disclosure;

[0017] Figure 3 is a schematic diagram according to a second embodiment of the present disclosure;

[0018] Figure 4 is a schematic diagram according to a third embodiment of the present disclosure;

[0019] Figure 5 is a schematic diagram according to a fourth embodiment of the present disclosure;

[0020] Figure 6 is a schematic diagram according to a fifth embodiment of the present disclosure;

[0021] Figure 7 is a schematic diagram according to a sixth embodiment of the present disclosure;

[0022] Figure 8 is a schematic diagram according to a seventh embodiment of the present disclosure;

[0023] Fig. 9is a schematic diagram according to an eighth embodiment of the present disclosure;

[0024] Fig.10 FIG. 1 is a schematic block diagram of an example electronic device 1000 that can be used to implement embodiments of the present disclosure. DETAILED DESCRIPTION

[0025] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0026] The present disclosure provides a method for determining a user experience threshold, a content distribution method and related devices, which are applied to technical fields such as content distribution in the field of artificial intelligence, the Internet and data processing, so as to achieve the purpose of improving the content distribution effect of multimedia.

[0027] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0028] Multimedia is one of the important means of promoting products and services, such as soft advertising. At present, the content distribution effect of multimedia is usually improved by adjusting the base price of unit multimedia traffic or adjusting the user experience constraints. However, the above methods are not effective in improving the content distribution effect of multimedia.

[0029] Taking multimedia as an example, in the current soft advertising scenario where user experience is required to be ensured, common problems include: (1) The challenge of improving content quality: Soft advertising platforms and advertisers (clients) have different goals for soft advertising content. Soft advertising platforms hope to ensure that user experience is within an acceptable range through high-quality soft advertising content and encourage clients to provide sustainable high-experience content; in contrast, clients are more concerned about the conversion efficiency of soft advertising to maximize revenue; however, at present, for clients, high-quality soft advertising materials usually mean higher production costs and weaker advertising properties, which may reduce their value; this also means that clients' understanding of soft advertising and content production capabilities still need time to iterate, and at the same time, it also means that soft advertising platforms need to use effective strategies to guide clients to actively improve the quality of their own content; (2) The challenge of traffic distribution and revenue optimization: In the case of limited traffic, soft advertising platforms need to improve content distribution effects while satisfying user experience constraints, thereby maximizing traffic revenue. To this end, soft advertising platforms must balance the relationship between traffic utilization, user experience constraints, and total traffic revenue.

[0030] Before allocating soft advertising traffic, the soft advertising platform usually limits the following two conditions: (1) Traffic quality: The soft advertising platform uses a model to score the impact of traffic on user experience. If the score is lower than a certain threshold, the corresponding content will be abandoned to ensure that the user experience is not affected; (2) The minimum price per unit of traffic: If the minimum price of traffic is too low, the corresponding content will also be abandoned to ensure the basic value of each traffic. It can be understood that the total traffic revenue satisfies the following formula: traffic * traffic utilization rate * unit traffic value. When traffic is limited, optimizing the total traffic revenue is mainly achieved by increasing traffic utilization rate and unit traffic value.

[0031] However, traffic utilization is affected by the floor price constraint and the user experience score threshold. Therefore, the optimization methods of total traffic revenue include: (1) lowering the floor price threshold constraint of unit traffic: that is, allowing unit traffic to be used at a lower price. Although this may increase traffic utilization, because the relaxation of restrictions allows more traffic to be utilized, it may also lead to a decrease in the value of unit traffic, resulting in a loss of total traffic revenue. Although the increase in traffic utilization may improve total traffic revenue, when the total amount of traffic is fixed, there are problems such as the lack of significant improvement in content distribution effect and limited room for improvement in total traffic revenue. (2) Lowering the user experience score threshold: By reducing the proportion of filtered content, traffic utilization can be improved. However, since the proportion of high-quality content in the overall content pool is limited, if the user experience threshold is too low, it may increase the distribution scale of low-quality content, bringing the risk of not meeting the user experience standard. Raising the threshold can ensure user experience, but it also leads to too much filtered content, resulting in low traffic utilization, which has a negative impact on total traffic revenue. Therefore, this method also has the problem of insufficient improvement in content distribution effect and limited optimization of total traffic revenue.

[0032] Under the current circumstances, improving the quality of customer content requires effective strategic incentives and sufficient time to deepen customers' understanding of soft advertising. The current common adjustment methods have very limited effects on improving the distribution effect of soft advertising and the increase in soft advertising revenue. This is also a key problem that the soft advertising distribution strategy of the current soft advertising platform urgently needs to solve. The improvement of soft advertising distribution effect mainly depends on adjusting the floor price of unit traffic or adjusting user experience constraints. However, since the total traffic of the soft advertising platform is limited and the soft advertising platform has strong user experience constraints, the improvement of soft advertising distribution effect is limited. From the perspective of optimizing user experience, the improvement of customer content production quality requires more effective guidance and incentives from soft advertising platforms.

[0033] In order to solve the above problems, the present disclosure provides a method for determining a user experience threshold, a content distribution method and related devices, which determine the user experience data corresponding to the content provider according to the content distribution posterior data and content distribution prior data corresponding to the content provider, and then dynamically determine the user experience threshold corresponding to the content provider based on the user experience data corresponding to the content provider, so that the user experience threshold is used to control the distribution of the content corresponding to the content provider, which can optimize the traffic allocation of content distribution, thereby optimizing the balance between content delivery and user experience, and effectively improving the content distribution effect; and based on the user experience data corresponding to the content provider, the content provider can be guided to provide higher quality content, thereby improving the overall content quality, and then improving the content distribution effect, and realizing the contribution of the content provider dimension to optimize the content distribution effect.

[0034] It can be seen that the method for determining the user experience threshold, the content distribution method and related devices provided by the present disclosure can be applied to determine the user experience threshold corresponding to the content provider (i.e., the customer, such as an advertiser), and then control the distribution of the corresponding content of the content provider according to the determined user experience threshold, such as soft advertisements.

[0035] Figure 1 Schematic diagram of an application scenario applicable to the present disclosure. Figure 1 As shown, the application scenario may include: a server cluster 11 and a terminal 12. The server cluster 11 includes multiple servers 111 and a memory 112, and the terminal 12 may be a tablet computer, a mobile phone, a laptop computer, a desktop computer, a smart home appliance, etc. The server 111 and the terminal 12 communicate via a wireless network or a wired network. The server 111 determines the user experience threshold corresponding to the content provider according to the method for determining the user experience threshold provided in the embodiment of the present disclosure, obtains the content to be distributed corresponding to the content provider from the material library according to the user experience threshold corresponding to the content provider, distributes the content to be distributed to the terminal 12, and displays the content to be distributed to the user through the terminal 12. In the above process, the memory 112 stores the relevant data of the server 111.

[0036] It should be noted that Figure 1 This is only a schematic diagram of an application scenario provided by the embodiment of the present disclosure. Figure 1 does not limit the equipment included in Figure 1 The positional relationship between the devices is limited.

[0037] The following specific embodiments are used to describe in detail the technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0038] Figure 2 Schematic diagram of the first embodiment of the present disclosure. Figure 2 As shown, the method for determining the user experience threshold provided in the first embodiment of the present disclosure can be applied to an electronic device, which may be a server or a server cluster, etc. The method for determining the user experience threshold provided in the first embodiment of the present disclosure includes:

[0039] S201: Obtain content distribution a posteriori data and content distribution a priori data corresponding to a target content provider.

[0040] In the disclosed embodiments, content distribution posterior data can be understood as user experience posterior scores obtained by users interacting with distributed content; content distribution prior data can be understood as user experience prior scores obtained by distributed content before distribution (i.e., before being presented to users); the target content provider is, for example, an advertiser, and the content provided by the target content provider is, for example, soft advertisements. Exemplarily, the content distribution posterior data of the distributed content can be determined based on the business indicators of the distributed content corresponding to the target content provider, and then the mean of the content distribution posterior data of the distributed content can be used as the content distribution posterior data corresponding to the target content provider. The content distribution prior data (i.e., a set of historical scores) of the distributed content can be obtained through a user experience prediction model, and then the mean of the content distribution prior data can be used as the content distribution prior data corresponding to the target content provider.

[0041] S202: Determine target user experience data corresponding to a target content provider according to the content distribution a priori data and the content distribution a priori data.

[0042] In this step, after obtaining the content distribution a posteriori data and the content distribution a priori data corresponding to the target content provider, the target user experience data corresponding to the target content provider can be determined based on the content distribution a posteriori data and the content distribution a priori data. The target user experience data can be understood as the total user experience score. For example, the target user experience data corresponding to the target content provider can be obtained by weighted summation based on the preset weights corresponding to the content distribution a posteriori data and the content distribution a priori data, so as to dynamically obtain the target user experience data corresponding to the target content provider. It can be understood that by distinguishing different user experience data (total user experience scores), a traffic reward and punishment mechanism based on the dimension of the target content provider can be implemented, that is, the target content providers that contribute to the overall user experience are rewarded, and the target content providers with poor experience are punished accordingly.

[0043] S203. Determine a user experience threshold corresponding to a target content provider according to the target user experience data and the user experience mean. The user experience threshold is used to control the distribution of content corresponding to the target content provider. The user experience mean is obtained based on the user experience data corresponding to the content provider.

[0044] Exemplarily, the user experience data corresponding to all content providers can be obtained, and then the average of the multiple user experience data can be obtained to obtain the user experience average. After determining the target user experience data corresponding to the target content provider, the target user experience data can be compared with the user experience average to determine the user experience threshold corresponding to the target content provider. For example, if the target user experience data is greater than the user experience average, the preset user experience threshold is lowered to obtain the lowered user experience threshold, and the lowered user experience threshold is used as the user experience threshold corresponding to the target content provider; if the target user experience data is less than or equal to the user experience average, the preset user experience threshold is increased to obtain the increased user experience threshold, and the increased user experience threshold is used as the user experience threshold corresponding to the target content provider, so as to dynamically determine the user experience threshold corresponding to the target content provider.

[0045] It can be understood that after each distribution of the content corresponding to the target content provider, the user experience threshold corresponding to the target content provider can be dynamically determined based on the distributed content corresponding to the target content provider by executing the above steps S201 to S203, which can be used to effectively control the distribution of the content corresponding to the target content provider, optimize the traffic allocation of content distribution, and improve the content distribution effect.

[0046] In the disclosed embodiment, target user experience data corresponding to the target content provider is determined based on content distribution posteriori data and content distribution priori data corresponding to the target content provider, and dynamic adjustment of the target user experience data is achieved to accurately determine the user experience threshold; the target user experience data is compared with the user experience mean to determine the user experience threshold corresponding to the target content provider, and dynamic adjustment of the user experience threshold is achieved. The user experience threshold is used to control the distribution of content corresponding to the target content provider, which can optimize the traffic distribution of content distribution, thereby optimizing the balance between content delivery and user experience, and effectively improving the content distribution effect; and based on the target user experience data corresponding to the target content provider, the target content provider can be guided to provide higher quality content, thereby improving the overall content quality, and then improving the content distribution effect, achieving a win-win situation for the content distribution platform and the target content provider.

[0047] Figure 3 FIG. 2 is a schematic diagram of a second embodiment of the present disclosure. Based on the above embodiment, the present disclosure further describes a method for determining a user experience threshold. Figure 3 As shown, the method for determining the user experience threshold provided by the second embodiment of the present disclosure may include:

[0048] In the present disclosure, Figure 2 The step S201 may further include the following two steps S301 and S302:

[0049] S301. Determine content distribution a posteriori data of the distributed content based on the business indicators of the distributed content corresponding to the target content provider; and obtain content distribution a posteriori data corresponding to the target content provider according to the content distribution a posteriori data of the distributed content.

[0050] Exemplarily, if the distributed content is a graphic type of content, the business indicators of the distributed content may include, for example, click-through rate, average reading time, and average reading depth, etc.; or, if the distributed content is a video type of content, the business indicators of the distributed content may include, for example, completion rate and fast scrolling rate, etc. The disclosed embodiments do not limit the business indicators. In this step, for example, the business indicators of the distributed content corresponding to the target content provider can be obtained based on the interaction log between the user and the distributed content, and then the content distribution posterior data of the distributed content can be determined based on the business indicators of the distributed content corresponding to the target content provider. It can be understood that if the quality of the distributed content is high, the obtained business indicators of the distributed content are high, so that the content distribution posterior data of the distributed content can be improved.

[0051] Optionally, the business indicator includes multiple business indicators, and determining the content distribution posterior data of the distributed content based on the business indicators of the distributed content corresponding to the target content provider may include: for each business indicator among the multiple business indicators, obtaining the ranking of the business indicator compared with the historical business indicator; based on the ranking, determining the indicator score corresponding to the business indicator; and performing weighted summation of the indicator scores to determine the content distribution posterior data of the distributed content.

[0052] Exemplarily, business indicators include click-through rate and dwell time, where the click-through rate is obtained based on whether the user clicks on the distributed content, and the dwell time is the length of time the user stays after clicking on the distributed content. Historical business indicators are based on the dimension of the content provider and are obtained based on the delivered content of different content providers. For example, taking the business indicator as the click-through rate as an example, the corresponding historical click-through rates include 4%, 6% and 10%, and the click-through rate of the distributed content corresponding to the target content provider is 11%, then the ranking of the click-through rate compared to the historical click-through rate can be obtained, that is, the ranking is the first, so that the indicator score corresponding to the click-through rate can be determined according to the ranking, such as the ranking is the first, and the corresponding indicator score is the highest score of 100 points. By analogy, the indicator scores corresponding to each business indicator can be determined, so that each indicator score can be summarized, and each indicator score can be weighted and summed to determine the content distribution posterior data of the distributed content, and store the content distribution posterior data of the distributed content. For example, the content distribution posterior data (user experience posterior score) of the distributed content can be stored in the user experience posterior score library of the content provider dimension.

[0053] S302, obtaining content distribution priori data of the distributed content corresponding to the target content provider; obtaining content distribution priori data corresponding to the target content provider according to the content distribution priori data of the distributed content.

[0054] Exemplarily, the mean of the content distribution priori data (user experience priori scores) of the distributed content may be obtained, and the mean is the content distribution priori data corresponding to the target content provider.

[0055] Optionally, obtaining content distribution prior data of distributed content corresponding to the target content provider may include: before the distributed content is distributed, obtaining content distribution prior data of the distributed content corresponding to the target content provider through a user experience prediction model, the user experience prediction model is trained based on distributed content samples corresponding to the content provider, distribution scenario information samples corresponding to the distributed content samples, and labels, and the labels include content distribution posterior data samples and business indicator samples of the distributed content.

[0056] Exemplarily, the user experience prediction model is obtained through pre-training, and can be regularly (regularly) updated and optimized based on actual content distribution posterior data. The training goal is to accurately predict the content distribution prior data of a new content before distributing it to users. For specific information on how to train the user experience prediction model, please refer to the subsequent embodiments, which will not be described in detail here. In this embodiment, before the distributed content is distributed, the content distribution prior data of the distributed content can be obtained through the user experience prediction model, and the content distribution prior data of the distributed content can be stored. For example, the content distribution prior data (user experience prior score) of the distributed content can be stored in the user experience prior score library of the content provider dimension.

[0057] It should be noted that the embodiment of the present disclosure does not limit the order in which steps S301 and S302 are executed.

[0058] In the present disclosure, Figure 2 The step S202 may further include the following step S303:

[0059] S303: Perform weighted summation on the content distribution a priori data and the content distribution a priori data to determine target user experience data corresponding to the target content provider.

[0060] Exemplarily, the weighted sum of the content distribution posterior data and the content distribution prior data is performed, that is, the content distribution posterior data and the content distribution prior data are converted in proportion, so as to determine the target user experience data corresponding to the target content provider. Specifically, the content distribution posterior data is the user experience posterior score, the content distribution prior data is the user experience prior score, and the target user experience data can be understood as the total user experience score. The target user experience data corresponding to the target content provider can be determined by the following formula:

[0061] α*user experience prior score + (1-α)*user experience posterior score = total user experience score

[0062] Among them, α∈(0,1), α represents a hyperparameter (i.e., preset weight).

[0063] Optionally, target user experience data corresponding to the target content provider may be output.

[0064] For example, after obtaining the target user experience data corresponding to the target content provider, the target user experience data can be output and displayed to the content provider, so that the content provider can intuitively perceive the current status of the provided content based on the account dimension and clearly understand its content quality. It can be understood that the display method of scores can avoid the tedious operation of measuring multiple related indicators when the content provider analyzes the account's own information, and directly display it to the content provider with the most intuitive numbers.

[0065] In the present disclosure, Figure 2 The step S203 may further include the following two steps S304 and S305:

[0066] S304. If the target user experience data is greater than the user experience mean, determine that the user experience threshold corresponding to the target content provider is a first threshold, where the first threshold is a threshold obtained by lowering a preset user experience threshold corresponding to the target content provider according to a first preset adjustment coefficient; the user experience threshold is used to control the distribution of content corresponding to the target content provider, and the user experience mean is obtained based on the user experience data corresponding to the content provider.

[0067] Exemplarily, the user experience data corresponding to the preset content providers (i.e., all content providers) can be obtained, and the average of the multiple user experience data can be obtained to obtain the user experience average. After determining the target user experience data corresponding to the target content provider, the target user experience data is compared with the user experience average. If the target user experience data is greater than the user experience average, the preset user experience threshold corresponding to the target content provider is lowered, such as multiplying the preset user experience threshold by the first preset coefficient to obtain the first threshold, thereby determining that the user experience threshold corresponding to the current target content provider is the first threshold. Among them, the preset user experience threshold can be understood as the default user experience threshold. By lowering the user experience threshold, the target content provider can enjoy a lower user experience threshold, thereby obtaining more exposure opportunities for high-value content, which helps to continuously motivate the target content provider to provide higher quality content, thereby improving the content delivery effect.

[0068] S305. If the target user experience data is less than or equal to the user experience mean, the user experience threshold corresponding to the target content provider is determined to be a second threshold, where the second threshold is a threshold obtained by increasing the preset user experience threshold according to a second preset adjustment coefficient; the user experience threshold is used to control the distribution of content corresponding to the target content provider, and the user experience mean is obtained based on the user experience data corresponding to the content provider.

[0069] Exemplarily, the target user experience data is compared with the user experience mean. If the target user experience data is less than or equal to the user experience mean, the preset user experience threshold corresponding to the target content provider is increased, such as multiplying the preset user experience threshold by the second preset coefficient to obtain the second threshold, thereby determining that the user experience threshold corresponding to the current target content provider is the second threshold. Among them, the second preset coefficient is greater than the first preset coefficient. By increasing the user experience threshold, the exposure of low-scoring content can be reduced, but the target content provider will face a higher user experience threshold, which can be understood as a punishment measure for the target content provider to guide the target content provider to improve the quality of the content.

[0070] It should be noted that the embodiment of the present disclosure does not limit the order of execution of steps S304 and S305. It is understood that after each distribution of the content corresponding to the target content provider, the user experience threshold corresponding to the target content provider can be dynamically determined by executing the above steps S301 to S305 based on the distributed content corresponding to the target content provider, so as to effectively control the distribution of the content corresponding to the target content provider and improve the content distribution effect.

[0071] In the disclosed embodiment, according to the content distribution a posteriori data of the distributed content corresponding to the target content provider, the content distribution a posteriori data corresponding to the target content provider is obtained; according to the content distribution a priori data of the distributed content corresponding to the target content provider, the content distribution a priori data corresponding to the target content provider is obtained; the content distribution a posteriori data and the content distribution a priori data are weighted and summed to determine the target user experience data corresponding to the target content provider, and the target user experience data is dynamically adjusted; if the target user experience data is greater than the user experience mean, the user experience threshold is lowered, so as to increase the exposure opportunity of the content corresponding to the target content provider, and encourage the target content provider to provide higher quality content, so as to effectively improve the content distribution effect; if the target user experience data is less than or equal to the user experience mean, the user experience threshold is increased to reduce the exposure of low-scoring content, ensure the content distribution effect, guide the target content provider to improve the quality of the content, and thus help improve the content distribution effect. The disclosed embodiment can effectively improve the content distribution effect and the user experience by dynamically adjusting the user experience threshold.

[0072] Based on the above embodiments, Figure 4 is a schematic diagram according to the third embodiment of the present disclosure. Figure 4 As shown, the training method of the user experience prediction model provided in the third embodiment of the present disclosure may include:

[0073] S401. Acquire training samples, where the training samples include distributed content samples corresponding to the content provider, distribution scenario information samples corresponding to the distributed content samples, and labels, where the labels include content distribution posterior data samples and business indicator samples of the distributed content.

[0074] Exemplarily, training samples can be obtained according to the interaction log of the distributed content corresponding to the user and the content provider. Among them, the interaction log may include, for example, display events, click events, and content distribution posterior data of each distributed content. Correspondingly, the training samples obtained according to the interaction log may include the distributed content samples corresponding to the content provider, the content distribution posterior data samples of the distributed content samples, and the business indicator samples. The distribution scene information samples corresponding to the distributed content samples are determined according to the distribution scenes of the distributed content samples. For example, the distribution scene information samples may include relevant information of the interactive users (i.e., user features), relevant information of the distributed content samples (i.e., content features), and relevant information of the scenes (i.e., environmental features); wherein, the relevant information of the interactive users may include, for example, user portrait related information such as gender, age, and activity, and the relevant information of the distributed content samples may include, for example, type (graphic type or video type), author related information, and classification information, and the relevant information of the scenes may include, for example, the time of interaction, the device used, and the column where the distributed content samples are displayed, such as the "discovery" column, the "recommendation" column, or the "video" column.

[0075] S402: Input the distributed content samples and the distribution scenario information samples into the user experience prediction model for training, so that the user experience prediction model fits the business indicator samples and obtains the predicted business indicator output by the user experience prediction model.

[0076] Exemplarily, the business indicator sample includes multiple business indicators, such as click-through rate samples and dwell time samples, etc. The distributed content samples and distribution scenario information samples can be input into the user experience prediction model for training, so that the user experience prediction model fits each business indicator sample, and obtains multiple predicted business indicators output by the user experience prediction model. For example, for the click-through rate, the training goal is to predict whether the user will click on the distributed content sample, as consistent as possible with the actual situation (i.e., the click-through rate sample); for the dwell time, the training goal is to predict the length of time the user stays after clicking on the distributed content sample, as consistent as possible with the actual situation (i.e., the dwell time sample). In addition to the click-through rate sample and the dwell time sample, other business indicator samples may also be included, such as the interaction rate, etc., and each has different situations according to the type of the distributed content sample. For example, for the distributed content sample of the video type, the corresponding business indicator sample may include the fast sliding rate (i.e., when swiping the video, a new video is swiped away immediately after a few seconds, and swiped to the next video). Among them, for click-through rate, since there are only two cases of click or no click, that is, click is 1 and no click is 0, for example, logarithmic loss can be used as the loss function, and the label is the click-through rate sample; for stay time, since it is a numerical estimate, for example, mean square error can be used as the loss function, and the label is the stay time sample.

[0077] S403: Based on the predicted business indicators, obtain predicted content distribution priori data.

[0078] For example, if the predicted business indicator is the predicted click rate, the predicted indicator score corresponding to the predicted click rate can be obtained according to the ranking of the predicted click rate. The predicted indicator scores corresponding to each predicted business indicator can be summarized, and the weighted sum of each predicted indicator score can be performed to obtain the predicted content distribution priori data.

[0079] S404: Adjust parameters of a user experience prediction model based on the predicted content distribution priori data and content distribution posterior data samples.

[0080] Exemplarily, the content distribution posterior data samples are labels, and the predicted content distribution prior data and the content distribution posterior data samples can be compared to determine the loss function value, thereby adjusting the parameters of the user experience prediction model according to the loss function value to obtain a trained user experience prediction model.

[0081] In the disclosed embodiment, a user experience prediction model is trained based on training samples so that the user experience prediction model fits the business indicator samples and accurately outputs the predicted business indicators; based on the predicted business indicators, predicted content distribution prior data is obtained, and based on the predicted content distribution prior data and content distribution posterior data samples of the distributed content samples, the parameters of the user experience prediction model are adjusted to obtain a trained user experience prediction model. Thus, when the user experience prediction model is used to obtain content distribution prior data corresponding to the content provider, the content distribution prior data of the content can be accurately output, which in turn helps to accurately determine the content to be distributed corresponding to the content provider and determine the user experience threshold corresponding to the content provider.

[0082] Based on the above embodiments, Figure 5 is a schematic diagram according to the fourth embodiment of the present disclosure. Figure 5 As shown, the content distribution method provided in the fifth embodiment of the present disclosure can be applied to an electronic device, which can be a server or a server cluster, etc. The content distribution method provided in the fourth embodiment of the present disclosure includes:

[0083] S501. Obtain the to-be-distributed content corresponding to the content provider from a material library according to the user experience threshold corresponding to the content provider, wherein the user experience threshold is determined by using the method for determining the user experience threshold in any of the above method embodiments.

[0084] In this step, after each distribution of the content to be distributed corresponding to the content provider, the user experience threshold is determined dynamically by using the method for determining the user experience threshold in any of the above method embodiments. Exemplarily, the content in the material library is provided by different content providers. For the content corresponding to different content providers in the material library, for example, the content distribution prior data (user experience prior score) of the content can be obtained through the user experience prediction model; the content distribution prior data of the content and the user experience threshold corresponding to the content provider are compared. If the content distribution prior data of the content is greater than the user experience threshold, the content is determined to be the content to be distributed corresponding to the content provider, that is, before the content is distributed, the content corresponding to the content provider in the material library is screened by the user experience threshold, and the content with higher quality is determined to be the content to be distributed. After determining that the content is the content to be distributed corresponding to the content provider, the content distribution prior data of the content to be distributed can be stored. For example, the content distribution prior data (user experience prior score) of the content to be distributed can be stored in the user experience prior score library of the content provider dimension, that is, the user experience prior score library is continuously updated through the content distribution prior data of the content to be distributed.

[0085] S502: Distribute the content to be distributed.

[0086] In this step, after obtaining the content to be distributed corresponding to the content provider, the content to be distributed can be distributed.

[0087] In the disclosed embodiment, according to the user experience threshold corresponding to the content provider, the content to be distributed corresponding to the content provider is obtained from the material library, and the content to be distributed is distributed. The user experience threshold is determined by using the user experience threshold determination method in any of the above method embodiments. According to the user experience threshold, high-quality content to be distributed can be accurately screened, thereby effectively improving the content distribution effect, optimizing the traffic distribution of content distribution, and further helping to improve the overall revenue, achieving a win-win situation for the content distribution platform and the content provider.

[0088] Figure 6 FIG. 5 is a schematic diagram of a fifth embodiment of the present disclosure. Based on the above embodiments, the present disclosure further describes the content distribution method. Figure 6 As shown, the content distribution method provided by the fifth embodiment of the present disclosure may include:

[0089] In the present disclosure, Figure 5 The step S501 may further include the following three steps S601 to S603:

[0090] S601. Obtain content corresponding to a content provider from a material library.

[0091] In this step, the contents in the material library are provided by different content providers, so the contents corresponding to the content providers can be obtained from the material library.

[0092] S602: Obtain content distribution priori data of the content.

[0093] In this step, for the content corresponding to the content provider, the content distribution prior data (user experience prior score) of the content can be obtained.

[0094] Furthermore, optionally, obtaining content distribution prior data of the content may include: inputting the content and the distribution scenario information corresponding to the content into a user experience prediction model to obtain content distribution prior data of the content, wherein the user experience prediction model is trained based on distributed content samples corresponding to the content provider, distribution scenario information samples corresponding to the distributed content samples, and labels, and the labels include content distribution posteriori data samples and business indicator samples of the distributed content.

[0095] Exemplarily, the distribution scenario information corresponding to the content is determined according to the distribution scenario of the content. For example, the distribution scenario information may include relevant information of the user to be interacted, relevant information of the content, and relevant information of the scenario; wherein, the relevant information of the user to be interacted may include, for example, user portrait related information such as gender, age, and activity, and the relevant information of the content may include, for example, type (graphic type or video type), author related information, and classification information, and the relevant information of the scenario may include, for example, the time when the interaction is to occur, the device to be used, and the column where the content is to be displayed, such as the "discovery" column, the "recommendation" column, or the "video" column. The user experience prediction model is trained using the training method of the user experience prediction model in any of the above method embodiments. By inputting the content and the distribution scenario information corresponding to the content into the user experience prediction model, the content distribution prior data of the content can be obtained.

[0096] S603: Determine that the content to be distributed is content whose prior distribution data is greater than a user experience threshold corresponding to the content provider.

[0097] In this step, the content distribution prior data of the content corresponding to the content provider is compared with the user experience threshold corresponding to the content provider. It can be determined that the content to be distributed is the content whose content distribution prior data is greater than the user experience threshold. That is, the content corresponding to the content provider is screened by the user experience threshold to obtain higher quality content to be distributed.

[0098] S604: Distribute the content to be distributed.

[0099] The implementation principle and technical effect of S604 can be referred to the above Figure 5 The embodiment of step S502 is not described in detail.

[0100] S605: Obtain target business indicators of the content to be distributed after distribution.

[0101] In this step, after distributing the content to be distributed, the target business indicators of the content to be distributed after distribution can be obtained by statistics based on the interaction log between the user and the content to be distributed. The target business indicators may include click-through rate, dwell time, like rate and conversion rate.

[0102] S606: Determine content distribution a posteriori data of the content to be distributed according to the target business indicator, and use the content distribution a posteriori data to train a user experience prediction model.

[0103] In this step, after obtaining the target service index of the content to be distributed after distribution, the content distribution posterior data of the content to be distributed can be determined according to the target service index.

[0104] Furthermore, the optional target business indicators include multiple ones, and determining the content distribution posterior data of the content to be distributed based on the target business indicators may include: for each target business indicator among the multiple target business indicators, obtaining the ranking of the target business indicator compared with the historical business indicators; determining the indicator score corresponding to the target business indicator based on the ranking; and performing weighted summation of the indicator scores to determine the content distribution posterior data of the content to be distributed.

[0105] Exemplarily, the target business indicators include, for example, click-through rate, dwell time, like rate and conversion rate, and the historical business indicators are obtained based on the dimensions of the content provider and the delivered content of different content providers. For each of the multiple target business indicators, the ranking of the target business indicator compared to the historical business indicator can be obtained, so as to determine the indicator score corresponding to the target business indicator based on the ranking, and then the scores of each indicator can be summarized, and the weighted sum of each indicator score can be performed to determine the content distribution posterior data of the content to be distributed, and the content distribution posterior data of the content to be distributed can be stored. For example, the content distribution posterior data of the content to be distributed can be stored in the user experience posterior score library of the content provider dimension, that is, the user experience posterior score library is continuously updated through the content distribution posterior data of the content to be distributed. The content distribution posterior data of the content to be distributed reflects the distribution effect of the content to be distributed among users, and can be used to generate the total user experience score corresponding to the content to be distributed, and then dynamically determine the user experience threshold corresponding to the content provider based on the total user experience score. The content distribution posterior data can be used to train the user experience prediction model. It is understandable that after each distribution of the content to be distributed, the user experience prediction model can be updated according to the content distribution a priori data of the content to be distributed to ensure that the user experience prediction model can accurately output the content distribution a priori data of the content in the material library.

[0106] S607: Obtain user experience data corresponding to the content to be distributed according to the content distribution priori data and content distribution a posteriori data of the content to be distributed, where the user experience data is used to adjust a user experience threshold corresponding to the content provider.

[0107] Exemplarily, referring to the example of step S303, the content distribution priori data and content distribution a posteriori data of the content to be distributed can be weighted and summed to obtain the user experience data (total user experience score) corresponding to the content to be distributed. After obtaining the user experience data corresponding to the content to be distributed, the user experience data can be used to adjust the user experience threshold corresponding to the content provider to dynamically determine the user experience threshold corresponding to the content provider.

[0108] In the disclosed embodiment, the content to be distributed is determined by screening the content corresponding to the content provider in the material library through the user experience threshold, which can ensure that the content to be distributed is of high quality and has a good content distribution effect; after distributing the content to be distributed, the content distribution posterior data of the content to be distributed can be determined based on the target business indicators of the content to be distributed after distribution, and the content distribution posterior data is used to train the user experience prediction model to achieve continuous optimization of the user experience prediction model; based on the content distribution a priori data and the content distribution posterior data of the content to be distributed, the user experience data corresponding to the content to be distributed is obtained, and the user experience data is used to adjust the user experience threshold corresponding to the content provider, so as to achieve dynamic determination of the user experience threshold corresponding to the content provider, thereby effectively improving the content distribution effect, and guiding the target content provider to improve the quality of the content, thereby helping to improve the overall revenue.

[0109] Based on the above embodiments, the method for determining the user experience threshold and the content distribution method provided by the embodiments of the present disclosure can ensure significant improvements in both user experience and revenue. According to the results of the current ab experiment (an experimental method for comparing two versions), the experimental group (a) is compared with the control group (b), while ensuring that the overall click-through rate is within the allowed constraints, the exposure of the experimental group (a) can be increased by, for example, about 17%, and the total traffic revenue of the content distribution platform can increase by, for example, 8%; at the same time, the traffic utilization rate can be increased by, for example, more than 95%, which can greatly optimize the resource utilization and revenue of the content distribution platform.

[0110] In summary, the technical solution provided by the present disclosure has at least the following advantages:

[0111] (1) It can optimize the balance between content (such as soft advertising) distribution and user experience: Based on the adjustment mechanism of the content provider's overall user experience data (total user experience score), the material dimension can be transformed into the account or content provider (customer) dimension. This adjustment can increase the flexibility of customers in obtaining material distribution opportunities. Specifically, content providers with high total user experience scores will enjoy a lower user experience threshold (user experience threshold), thereby obtaining more exposure opportunities for high-value content; while content providers with low total user experience scores will face a higher user experience threshold. The above method is similar to adjusting the evaluation criteria to only require an average score of, for example, 90 points, thereby giving content providers more opportunities to participate while ensuring the interests of the content distribution platform (such as the soft advertising platform), while optimizing the balance between content distribution and user experience.

[0112] (2) It can ensure the sustainable distribution of high-quality content: Implement incentive programs for content providers whose user experience data (total user experience score) is above the average level to encourage them to submit higher-quality content. This can not only improve the overall content quality, but also provide the content provider with more high-value content exposure opportunities, which helps to continuously motivate the production of high-quality content.

[0113] (3) The present disclosure provides a win-win feedback mechanism and productization mechanism: by lowering the publishing threshold for content providers corresponding to high-quality content, a win-win feedback mechanism can be established. While content providers gain more exposure opportunities, content distribution platforms can achieve greater traffic utilization; a productization mechanism for experience score presentation and feedback is provided, so that content providers can clearly understand their content quality and market positioning and reap higher profits, while content distribution platforms can also obtain better user experience and value;

[0114] (4) Mechanism perceptibility: It can monitor and analyze user interaction data in real time to dynamically adjust content distribution strategies. It can automatically detect changes in user experience data (total user experience score) through advanced algorithms and flexibly adjust the priority of content exposure based on the changes. For example, when the total user experience score of a certain content increases, it can automatically lower its exposure threshold and increase its exposure opportunities. When the total user experience score of a certain content decreases, it can raise the threshold and reduce the exposure of low-scoring content. The above mechanism can ensure that the content distribution platform can quickly respond to user feedback, maintain the healthy development of the content ecosystem, and improve overall user satisfaction. This perceptibility can not only optimize the user experience, but also provide the content distribution platform with more accurate content management tools, helping the content distribution platform and content providers achieve greater success in content distribution.

[0115] (5) Intuitive perception of account status: By introducing a scoring mechanism, the status of the account is evaluated and displayed, and the score enables the content provider to intuitively perceive the current status. The score presentation method can avoid the tedious operation of measuring multiple related indicators when analyzing the account's own information, and directly present it to the content provider with the most intuitive numbers. At the same time, the content provider can also obtain the content distribution platform's preference for each indicator through the score corresponding to the account, which is conducive to the content provider to optimize in a direction that is more in line with the wishes of the content distribution platform after understanding the score corresponding to the account.

[0116] Through refined design, the present disclosure can realize the sharing of the benefits brought by the supply of high-quality content, and can effectively share the user experience losses caused by content providers (customers). By introducing a dynamic scoring mechanism, the feedback method can be concretized, further enhancing customers' perception of their own competitive relationships and current positioning. In addition, the scoring mechanism can motivate customers to invest more time and energy in providing high-quality content, thereby helping to drive the content distribution platform to rationally distribute the traffic of high-quality content, thereby improving the content distribution effect.

[0117] Figure 7 is a schematic diagram according to the sixth embodiment of the present disclosure. Figure 7 As shown, the device 700 for determining the user experience threshold provided by the sixth embodiment of the present disclosure includes: an acquisition unit 701, used to acquire content distribution posteriori data and content distribution priori data corresponding to the target content provider; a first determination unit 702, used to determine the target user experience data corresponding to the target content provider based on the content distribution posteriori data and the content distribution priori data; a second determination unit 703, used to determine the user experience threshold corresponding to the target content provider based on the target user experience data and the user experience mean, the user experience threshold is used to control the distribution of content corresponding to the target content provider, and the user experience mean is obtained based on the user experience data corresponding to the content provider.

[0118] In some embodiments, the second determination unit 703 includes: a first threshold determination module (not shown in the figure), which is used to determine that the user experience threshold corresponding to the target content provider is a first threshold if the target user experience data is greater than the user experience mean, and the first threshold is a threshold obtained by lowering the preset user experience threshold corresponding to the target content provider according to the first preset adjustment coefficient; a second threshold determination module (not shown in the figure), which is used to determine that the user experience threshold corresponding to the target content provider is a second threshold if the target user experience data is less than or equal to the user experience mean, and the second threshold is a threshold obtained by increasing the preset user experience threshold according to the second preset adjustment coefficient.

[0119] In some embodiments, the acquisition unit 701 includes: a content distribution a posteriori data determination module (not shown in the figure), which is used to determine the content distribution a posteriori data of the distributed content based on the business indicators of the distributed content corresponding to the target content provider; a content distribution a posteriori data acquisition module (not shown in the figure), which is used to acquire the content distribution a posteriori data corresponding to the target content provider based on the content distribution a posteriori data of the distributed content.

[0120] In some embodiments, the business indicator includes multiple ones, and the content distribution posterior data determination module includes: a content distribution posterior data acquisition submodule (not shown in the figure), which is used to obtain the ranking of the business indicator compared with the historical business indicator for each business indicator among the multiple business indicators; a first determination submodule (not shown in the figure), which is used to determine the indicator score corresponding to the business indicator based on the ranking; and a second determination submodule (not shown in the figure), which is used to perform weighted summation of each indicator score to determine the content distribution posterior data of the distributed content.

[0121] In some embodiments, the acquisition unit 701 includes: a first content distribution prior data acquisition module (not shown in the figure), which is used to obtain content distribution prior data of the distributed content corresponding to the target content provider; a second content distribution prior data acquisition module (not shown in the figure), which is used to obtain content distribution prior data corresponding to the target content provider based on the content distribution prior data of the distributed content.

[0122] In some embodiments, the first content distribution prior data acquisition module includes: a content distribution prior data acquisition sub-module (not shown in the figure), which is used to obtain content distribution prior data of the distributed content through a user experience prediction model before the distributed content is distributed. The user experience prediction model is trained based on the distributed content samples corresponding to the content provider, the distribution scenario information samples corresponding to the distributed content samples, and labels, and the labels include content distribution posterior data samples and business indicator samples of the distributed content.

[0123] In some embodiments, the user experience threshold determination device 700 also includes a training unit (not shown in the figure), which is used to train and obtain a user experience prediction model in the following manner: obtaining training samples, the training samples include distributed content samples corresponding to the content provider, distribution scenario information samples corresponding to the distributed content samples, and labels, and the labels include content distribution posterior data samples and business indicator samples of the distributed content samples; inputting the distributed content samples and the distribution scenario information samples into the user experience prediction model for training, so that the user experience prediction model fits the business indicator samples, and obtains the predicted business indicators output by the user experience prediction model; based on the predicted business indicators, obtaining predicted content distribution prior data; based on the predicted content distribution prior data and the content distribution posterior data samples, adjusting the parameters of the user experience prediction model.

[0124] In some embodiments, the first determination unit 702 includes: a user experience data determination module (not shown in the figure), which is used to perform weighted summation on content distribution a posteriori data and content distribution a priori data to determine target user experience data corresponding to the target content provider.

[0125] Figure 7 The provided device for determining a user experience threshold can execute the steps in the method embodiment corresponding to the above-mentioned method for determining a user experience threshold. Its implementation principle and technical effects are similar and will not be repeated here.

[0126] Figure 8 is a schematic diagram according to the seventh embodiment of the present disclosure. Figure 8 As shown, the content distribution device 800 provided in the seventh embodiment of the present disclosure includes: an acquisition unit 801, used to acquire the content to be distributed corresponding to the content provider from the material library according to the user experience threshold corresponding to the content provider, wherein the user experience threshold is determined by using the method for determining the user experience threshold in any of the above method embodiments; a distribution unit 802, used to distribute the content to be distributed.

[0127] In some embodiments, the acquisition unit 801 includes: a content acquisition module (not shown in the figure), which is used to acquire content corresponding to a content provider from a material library; a content distribution prior data acquisition module (not shown in the figure), which is used to acquire content distribution prior data of the content; and a determination module (not shown in the figure), which is used to determine that the content to be distributed is content whose content distribution prior data is greater than a user experience threshold.

[0128] In some embodiments, the content distribution prior data acquisition module includes: a content distribution prior data acquisition sub-module (not shown in the figure), which is used to input the content and the distribution scenario information corresponding to the content into a user experience prediction model to obtain the content distribution prior data of the content. The user experience prediction model is trained based on the distributed content samples corresponding to the content provider, the distribution scenario information samples corresponding to the distributed content samples, and labels. The labels include content distribution posterior data samples and business indicator samples of the distributed content.

[0129] In some embodiments, the content distribution device 800 also includes a training unit (not shown in the figure), which is used to train and obtain a user experience prediction model in the following manner: obtaining training samples, the training samples include distributed content samples corresponding to the content provider, distribution scenario information samples corresponding to the distributed content samples, and labels, and the labels include content distribution posterior data samples and business indicator samples of the distributed content samples; inputting the distributed content samples and the distribution scenario information samples into the user experience prediction model for training, so that the user experience prediction model fits the business indicator samples, and obtains the predicted business indicators output by the user experience prediction model; based on the predicted business indicators, obtaining predicted content distribution prior data; based on the predicted content distribution prior data and the content distribution posterior data samples, adjusting the parameters of the user experience prediction model.

[0130] In some embodiments, the content distribution device 800 also includes a processing unit (not shown in the figure), which is used to: obtain target business indicators of the content to be distributed after distribution; determine content distribution posterior data of the content to be distributed based on the target business indicators, and the content distribution posterior data is used to train a user experience prediction model; obtain user experience data corresponding to the content to be distributed based on the content distribution prior data and content distribution posterior data of the content to be distributed, and the user experience data is used to adjust the user experience threshold corresponding to the content provider.

[0131] In some embodiments, the target business indicator includes multiple ones, and the processing unit includes: an acquisition module (not shown in the figure), which is used to obtain the ranking of the target business indicator compared with the historical business indicators for each target business indicator among the multiple target business indicators; a first determination module (not shown in the figure), which is used to determine the indicator score corresponding to the target business indicator according to the ranking; and a second determination module (not shown in the figure), which is used to perform weighted summation of each indicator score to determine the content distribution posterior data of the content to be distributed.

[0132] Figure 8 The provided content distribution device can execute the steps in the method embodiment corresponding to the above-mentioned content distribution method. Its implementation principle and technical effects are similar and will not be repeated here.

[0133] Fig. 9 is a schematic diagram according to the eighth embodiment of the present disclosure. Fig. 9 As shown, based on Figure 7 A device for determining a user experience threshold is provided and Figure 8The content distribution device provided, the content provider (customer) uploads the corresponding content to the material library through the content distribution platform (supply content), the content distribution device of the content distribution platform obtains the content to be distributed corresponding to the content provider from the material library according to the user experience threshold corresponding to the content provider, and distributes the content to be distributed, wherein the content distribution priori data of the content to be distributed (for example, obtained through a user experience prediction model) is greater than the user experience threshold; the content distribution posterior data (user experience posterior score) of the content to be distributed is obtained and updated to the user experience posterior score library of the content provider (customer) dimension, and the content distribution posterior data is used to optimize the user experience prediction model. The device for determining the user experience threshold of the content distribution platform obtains the content distribution posterior data (user experience posterior score) and content distribution a priori data (user experience a priori score) corresponding to the content provider; determines the user experience data (total user experience score) corresponding to the content provider based on the content distribution posterior data and the content distribution a priori data, and the user experience data is used for reference to adjust the user experience threshold corresponding to the content provider; determines the user experience threshold corresponding to the content provider based on the user experience data and the user experience mean, and the user experience threshold is used to control the distribution of the content corresponding to the content provider, and the user experience mean is obtained based on the preset user experience data corresponding to the content provider.

[0134] It can be understood that through the synergy of the user experience threshold determination device and the content distribution device, the multimedia content distribution effect and user experience can be effectively improved. The user experience threshold determination device dynamically adjusts the content provider's total user experience score based on the content distribution posterior data and content distribution prior data of the content to ensure the accuracy of the evaluation. The content distribution device adjusts the distribution of content (such as soft advertisements) according to the total user experience score, and encourages content providers to provide high-quality content. Through the collaborative work of the two devices, the content distribution platform can optimize traffic distribution and increase content (such as soft advertisements) revenue while ensuring user experience, thereby achieving a win-win situation for the content distribution platform and content providers (customers).

[0135] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device, the electronic device comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the scheme provided in any of the above embodiments. According to an embodiment of the present disclosure, the present disclosure further provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute the scheme provided in any of the above embodiments. According to an embodiment of the present disclosure, the present disclosure further provides a computer program product, the computer program product comprising: a computer program, the computer program is stored in a readable storage medium, at least one processor of the electronic device can read the computer program from the readable storage medium, and at least one processor executes the computer program so that the electronic device executes the scheme provided in any of the above embodiments.

[0136] Fig.10 1 is a schematic block diagram of an example electronic device 1000 that can be used to implement an embodiment of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0137] like Fig.10 As shown, the electronic device 1000 includes a computing unit 1001, which can store data stored in a read-only memory (ROM) ( Fig.10 A computer program in ROM 1002 is loaded from storage unit 1008 to random access memory (RAM) ( Fig.10 The computer programs in the RAM 1003 are used to perform various appropriate actions and processes. The RAM 1003 may also store various programs and data required for the operation of the electronic device 1000. The computing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. The input / output (I / O) interface ( Fig.10 The I / O interface 1005 is also connected to the bus 1004 .

[0138] Multiple components in the electronic device 1000 are connected to the I / O interface 1005, including: an input unit 1006, such as a keyboard, a mouse, etc.; an output unit 1007, such as various types of displays, speakers, etc.; a storage unit 1008, such as a disk, an optical disk, etc.; and a communication unit 1009, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1009 allows the electronic device 1000 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0139] The computing unit 1001 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSP), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1001 performs the various methods and processes described above, such as the content distribution method. For example, in some embodiments, the content distribution method may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 1008. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 1000 via the ROM 1002 and / or the communication unit 1009. When the computer program is loaded into the RAM 1003 and executed by the computing unit 1001, one or more steps of the content distribution method described above may be performed. Alternatively, in other embodiments, the computing unit 1001 may be configured to execute the content distribution method in any other appropriate manner (eg, by means of firmware).

[0140] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard parts (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0141] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0142] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0143] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0144] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: Local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

[0145] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server may also be a server of a distributed system, or a server combined with a blockchain.

[0146] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0147] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, fusions, sub-fusions and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure should be included in the protection scope of the present disclosure.

Claims

1. A method for determining a user experience threshold, comprising: Obtaining content distribution a posteriori data and content distribution a priori data corresponding to the target content provider; Determining target user experience data corresponding to the target content provider according to the content distribution a posteriori data and the content distribution a priori data; A user experience threshold corresponding to the target content provider is determined based on the target user experience data and the user experience mean. The user experience threshold is used to control the distribution of content corresponding to the target content provider. The user experience mean is obtained based on the user experience data corresponding to the content provider.

2. The determination method according to claim 1, wherein: The determining, according to the target user experience data and the user experience mean, a user experience threshold corresponding to the target content provider includes: If the target user experience data is greater than the user experience mean, determining that the user experience threshold corresponding to the target content provider is a first threshold, where the first threshold is a threshold obtained by lowering a preset user experience threshold corresponding to the target content provider according to a first preset adjustment coefficient; If the target user experience data is less than or equal to the user experience mean, the user experience threshold corresponding to the target content provider is determined to be a second threshold, where the second threshold is a threshold obtained by increasing the preset user experience threshold according to a second preset adjustment coefficient.

3. The determination method according to claim 1 or 2, wherein: The obtaining of content distribution posterior data corresponding to the target content provider includes: Determining content distribution posterior data of the distributed content based on the business indicators of the distributed content corresponding to the target content provider; According to the content distribution a posteriori data of the distributed content, the content distribution a posteriori data corresponding to the target content provider is obtained.

4. The determination method according to claim 3, wherein: The business indicator includes a plurality of business indicators, and determining the content distribution posterior data of the distributed content based on the business indicator of the distributed content corresponding to the target content provider includes: For each of the multiple business indicators, obtain a ranking of the business indicator compared to historical business indicators; Determining, according to the ranking, an indicator score corresponding to the business indicator; A weighted sum is taken for each of the indicator scores to determine content distribution posterior data of the distributed content.

5. The determination method according to claim 1 or 2, wherein: The obtaining of the content distribution priori data corresponding to the target content provider includes: Acquire content distribution priori data of the distributed content corresponding to the target content provider; According to the content distribution priori data of the distributed content, the content distribution priori data corresponding to the target content provider is obtained.

6. The determination method according to claim 5, wherein: The obtaining of content distribution priori data of the distributed content corresponding to the target content provider includes: Before the distributed content is distributed, content distribution priori data of the distributed content is obtained through a user experience prediction model, and the user experience prediction model is trained based on the distributed content samples corresponding to the content provider, the distribution scenario information samples corresponding to the distributed content samples, and labels, and the labels include content distribution posteriori data samples and business indicator samples of the distributed content.

7. The determination method according to claim 6, wherein: The user experience prediction model is trained in the following way: Acquire training samples, wherein the training samples include distributed content samples corresponding to the content provider, distribution scenario information samples corresponding to the distributed content samples, and labels, wherein the labels include content distribution posterior data samples and business indicator samples of the distributed content samples; Inputting the distributed content sample and the distribution scenario information sample into a user experience prediction model for training, so that the user experience prediction model fits the business indicator sample, and obtaining the predicted business indicator output by the user experience prediction model; Based on the predicted business indicator, obtaining predicted content distribution priori data; Based on the predicted content distribution priori data and the content distribution posterior data samples, the parameters of the user experience prediction model are adjusted.

8. The determination method according to any one of claims 1 to 7, wherein: The determining, according to the content distribution a priori data and the content distribution a priori data, target user experience data corresponding to the target content provider comprises: A weighted sum is performed on the content distribution a priori data and the content distribution a priori data to determine target user experience data corresponding to the target content provider.

9. A content distribution method, comprising: According to the user experience threshold corresponding to the content provider, obtaining the to-be-distributed content corresponding to the content provider from the material library, wherein the user experience threshold is determined by using the method for determining the user experience threshold according to any one of claims 1 to 8; Distribute the content to be distributed.

10. The content distribution method according to claim 9, wherein: The acquiring, from a material library, the to-be-distributed content corresponding to the content provider according to the user experience threshold corresponding to the content provider, comprises: Acquire the content corresponding to the content provider from the material library; Acquire content distribution priori data of the content; The to-be-distributed content is determined to be content whose content distribution priori data is greater than the user experience threshold.

11. The content distribution method according to claim 10, wherein: The obtaining of content distribution priori data of the content includes: The content and the distribution scenario information corresponding to the content are input into a user experience prediction model to obtain content distribution priori data of the content. The user experience prediction model is trained based on the distributed content samples corresponding to the content provider, the distribution scenario information samples corresponding to the distributed content samples, and labels, and the labels include content distribution posteriori data samples and business indicator samples of the distributed content.

12. The content distribution method according to claim 11, wherein: The user experience prediction model is trained in the following way: Acquire training samples, wherein the training samples include distributed content samples corresponding to the content provider, distribution scenario information samples corresponding to the distributed content samples, and labels, wherein the labels include content distribution posterior data samples and business indicator samples of the distributed content samples; Inputting the distributed content sample and the distribution scenario information sample into a user experience prediction model for training, so that the user experience prediction model fits the business indicator sample, and obtaining the predicted business indicator output by the user experience prediction model; Based on the predicted business indicator, obtaining predicted content distribution priori data; Based on the predicted content distribution priori data and the content distribution posterior data samples, the parameters of the user experience prediction model are adjusted.

13. The content distribution method according to any one of claims 9 to 12, wherein: The method further comprises: Obtaining target business indicators of the content to be distributed after distribution; Determining content distribution a posteriori data of the content to be distributed according to the target business indicator, wherein the content distribution a posteriori data is used to train a user experience prediction model; According to the content distribution a priori data and the content distribution a posteriori data of the content to be distributed, user experience data corresponding to the content to be distributed is obtained, and the user experience data is used to adjust the user experience threshold corresponding to the content provider.

14. The content distribution method according to claim 13, wherein: The target business indicator includes a plurality of indicators, and determining the content distribution posterior data of the content to be distributed according to the target business indicator includes: For each target business indicator among the plurality of target business indicators, obtaining a ranking of the target business indicator compared with historical business indicators; Determining, according to the ranking, an indicator score corresponding to the target business indicator; A weighted sum is taken for each of the indicator scores to determine content distribution posterior data of the content to be distributed.

15. A device for determining a user experience threshold, comprising: An acquisition unit, used to acquire content distribution a posteriori data and content distribution a priori data corresponding to a target content provider; A first determining unit, configured to determine target user experience data corresponding to the target content provider according to the content distribution a posteriori data and the content distribution a priori data; The second determination unit is used to determine the user experience threshold corresponding to the target content provider based on the target user experience data and the user experience mean, wherein the user experience threshold is used to control the distribution of content corresponding to the target content provider, and the user experience mean is obtained based on the user experience data corresponding to the content provider.

16. The determination device according to claim 15, wherein: The second determining unit includes: A first threshold determination module, configured to determine, if the target user experience data is greater than the user experience mean, that the user experience threshold corresponding to the target content provider is a first threshold, wherein the first threshold is a threshold obtained by lowering a preset user experience threshold corresponding to the target content provider according to a first preset adjustment coefficient; The second threshold determination module is used to determine that the user experience threshold corresponding to the target content provider is a second threshold if the target user experience data is less than or equal to the user experience mean, and the second threshold is a threshold obtained by increasing the preset user experience threshold according to a second preset adjustment coefficient.

17. The determination device according to claim 15 or 16, wherein: The acquisition unit comprises: A content distribution posterior data determination module, configured to determine content distribution posterior data of the distributed content based on the business indicators of the distributed content corresponding to the target content provider; The content distribution a posteriori data acquisition module is used to acquire the content distribution a posteriori data corresponding to the target content provider according to the content distribution a posteriori data of the distributed content.

18. The determination device according to claim 17, wherein: The business indicators include multiple ones, and the content distribution posterior data determination module includes: A content distribution posterior data acquisition submodule, used for acquiring, for each of the multiple business indicators, a ranking of the business indicator compared with the historical business indicators; A first determination submodule, configured to determine an indicator score corresponding to the business indicator according to the ranking; The second determination submodule is configured to perform a weighted summation on the index scores to determine the content distribution posterior data of the distributed content.

19. The determination device according to claim 15 or 16, wherein: The acquisition unit comprises: A first content distribution priori data acquisition module, used to acquire content distribution priori data of the distributed content corresponding to the target content provider; The second content distribution priori data acquisition module is used to acquire content distribution priori data corresponding to a target content provider according to the content distribution priori data of the distributed content.

20. The determination device according to claim 19, wherein: The first content distribution priori data acquisition module includes: A content distribution prior data acquisition submodule is used to obtain content distribution prior data of the distributed content through a user experience prediction model before the distributed content is distributed, wherein the user experience prediction model is trained based on the distributed content samples corresponding to the content provider, the distribution scenario information samples corresponding to the distributed content samples, and labels, and the labels include content distribution posterior data samples and business indicator samples of the distributed content.

21. The determination device according to claim 20, wherein: It also includes a training unit, which is used to train and obtain the user experience prediction model in the following manner: obtaining training samples, the training samples including distributed content samples corresponding to the content provider, distribution scenario information samples corresponding to the distributed content samples, and labels, the labels including content distribution posterior data samples and business indicator samples of the distributed content samples; inputting the distributed content samples and the distribution scenario information samples into the user experience prediction model for training, so that the user experience prediction model fits the business indicator samples to obtain the predicted business indicators output by the user experience prediction model; based on the predicted business indicators, obtaining predicted content distribution prior data; based on the predicted content distribution prior data and the content distribution posterior data samples, adjusting the parameters of the user experience prediction model.

22. The determination device according to any one of claims 15 to 21, wherein: The first determining unit includes: The user experience data determination module is used to perform weighted summation on the content distribution a priori data and the content distribution a priori data to determine the target user experience data corresponding to the target content provider.

23. A content distribution device, comprising: an acquisition unit, configured to acquire, from a material library, content to be distributed corresponding to a content provider according to a user experience threshold corresponding to the content provider, wherein the user experience threshold is determined by using the method for determining the user experience threshold according to any one of claims 1 to 8; A distribution unit is used to distribute the content to be distributed.

24. The content distribution device according to claim 23, wherein: The acquisition unit comprises: A content acquisition module, used to acquire the content corresponding to the content provider from the material library; A content distribution priori data acquisition module, used to acquire content distribution priori data of the content; The determination module is used to determine that the content to be distributed is content whose content distribution priori data is greater than the user experience threshold.

25. The content distribution device according to claim 24, wherein: The content distribution priori data acquisition module includes: The content distribution prior data acquisition submodule is used to input the content and the distribution scenario information corresponding to the content into a user experience prediction model to obtain the content distribution prior data of the content. The user experience prediction model is trained based on the distributed content samples corresponding to the content provider, the distribution scenario information samples corresponding to the distributed content samples, and labels. The labels include content distribution posterior data samples and business indicator samples of the distributed content.

26. The content distribution device according to claim 25, wherein: It also includes a training unit, which is used to train and obtain the user experience prediction model in the following manner: obtaining training samples, the training samples including distributed content samples corresponding to the content provider, distribution scenario information samples corresponding to the distributed content samples, and labels, the labels including content distribution posterior data samples and business indicator samples of the distributed content samples; inputting the distributed content samples and the distribution scenario information samples into the user experience prediction model for training, so that the user experience prediction model fits the business indicator samples to obtain the predicted business indicators output by the user experience prediction model; based on the predicted business indicators, obtaining predicted content distribution prior data; based on the predicted content distribution prior data and the content distribution posterior data samples, adjusting the parameters of the user experience prediction model.

27. The content distribution device according to any one of claims 23 to 26, wherein: It also includes a processing unit, which is used to: obtain the target business indicator of the content to be distributed after distribution; determine the content distribution posterior data of the content to be distributed according to the target business indicator, and the content distribution posterior data is used to train the user experience prediction model; obtain the user experience data corresponding to the content to be distributed according to the content distribution priori data of the content to be distributed and the content distribution posterior data, and the user experience data is used to adjust the user experience threshold corresponding to the content provider.

28. The content distribution device according to claim 27, wherein: The target business indicators include multiple ones, and the processing unit includes: An acquisition module, configured to acquire, for each of the plurality of target business indicators, a ranking of the target business indicator compared with historical business indicators; A first determination module, configured to determine an indicator score corresponding to the target business indicator according to the ranking; The second determination module is used to perform weighted summation on the index scores to determine the content distribution posterior data of the content to be distributed.

29. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method for determining the user experience threshold according to any one of claims 1 to 8 or the content distribution method according to any one of claims 9 to 14.

30. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to enable the computer to execute the method for determining a user experience threshold according to any one of claims 1 to 8 or the content distribution method according to any one of claims 9 to 14.

31. A computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the method for determining a user experience threshold according to any one of claims 1 to 8 or the steps of the content distribution method according to any one of claims 9 to 14.

Images