Material delivery method and device, computer program product and electronic equipment

Abstract

The invention relates to the technical field of computers, in particular to a material delivery method and device, a computer program product and electronic equipment. The material putting method comprises the steps that user feedback features of put materials are collected, first reward data of the put materials are determined based on the user feedback features, the first reward data reflect the putting effect of the put materials, and the put materials are generated based on at least one material segment obtained from a material library; performing material attribution processing according to the second reward data to determine the second reward data of the material segment; and updating a material effect score of the material fragment according to the second reward data and a historical material index of the material fragment, and obtaining a target material fragment from a material library based on the updated material effect score to generate a target material and putting the target material. According to the invention, the quality and efficiency of material putting can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and more specifically, to a material delivery method, a material delivery device, a computer program product, and an electronic device. Background Art

[0002] With the development of computer and Internet technology, the delivery of materials to various platforms based on the Internet has the characteristics of instantaneity and wide coverage. For example, videos, audios, texts, pictures and combinations of various types of materials that users are interested in can be delivered on web pages and applications. However, the delivery of a material involves steps such as material selection, editing, and editing optimization. It cannot capture the user's interests and respond to changes in user needs, which affects the quality and efficiency of material delivery to a certain extent.

[0003] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0004] The purpose of the present disclosure is to provide a material delivery method and device, a computer program product and an electronic device, thereby improving the quality and efficiency of delivered materials at least to a certain extent.

[0005] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0006] According to one aspect of the present disclosure, a material delivery method is provided, comprising: collecting user feedback features of a delivered video, and determining first reward data for the delivered material based on the user feedback features, wherein the first reward data reflects delivery effect of the delivered material, and the delivered material is generated based on at least one material clip obtained from a material library; performing material attribution processing based on the first reward data to determine second reward data for the material clip; updating a material effect score of the material clip based on the second reward data and a historical material indicator of the material clip, and acquiring a target material clip from the material library based on the updated material effect score to generate a target material and deliver the target material.

[0007] In an exemplary embodiment of the present disclosure, determining the first reward data of the released material based on the user feedback features includes: obtaining weight information corresponding to the user feedback features, the weight information reflecting the importance of the user feedback features in the first reward data; based on the weight information, fusing the user feedback features corresponding to the released material to obtain the first reward data.

[0008] In an exemplary embodiment of the present disclosure, the material attribution processing is performed according to the first reward data to determine the second reward data of the material fragment, including: determining the material weight corresponding to the material fragment according to the fragment position of the material fragment in the released material; and determining the second reward data of the material fragment based on the material weight and the first reward data.

[0009] In an exemplary embodiment of the present disclosure, determining the material weight corresponding to the material fragment according to the fragment position of the material fragment in the already-released material includes: determining the material weight corresponding to the material fragment according to the fragment position based on a preset weight decreasing model; wherein the weight decreasing model is used to characterize the change of the material weight with the fragment position.

[0010] In an exemplary embodiment of the present disclosure, updating the material effect score of the material clip according to the second reward data and the historical material indicators of the material clip includes: obtaining a current material reward corresponding to the material clip; obtaining a reward difference value between the current material reward and the second reward data; determining a reward correction parameter based on the historical material indicators, and updating the material effect score of the material clip according to the reward correction parameter, the reward difference value and the current material reward.

[0011] In an exemplary embodiment of the present disclosure, the historical material indicator includes at least one of the number of times the material segment is selected, the number of times the material is completed, and the number of times the material is interacted; wherein the number of times the material segment is selected indicates the number of times the material segment is used to generate historically released materials, the number of times the material is completed indicates the number of times the historically released materials are completely played, and the number of material interactions indicates the total number of interactions that occurred with the historically released materials during the playback of the material segment.

[0012] In an exemplary embodiment of the present disclosure, the target material fragment is obtained from the material library based on the updated material effect score to generate the target material and deliver it, including: obtaining the material exploration probability, the material exploration probability is used to indicate the proportion of random material in the target material fragment; obtaining random material from the material library according to the material exploration probability, and obtaining generated material from the material library based on the updated material effect score, so as to determine the target material fragment according to the random material and the generated material and generate the target material for delivery.

[0013] In an exemplary embodiment of the present disclosure, material clips in a material library have respective corresponding category labels; based on updated material effect scores, target material clips are obtained from the material library to generate target materials and then delivered, including: based on updated material effect scores, target material clips are obtained from the material library to generate multiple materials to be delivered; for each material to be delivered, the material compatibility of the material to be delivered is evaluated according to the category label of the target material clip included in the material to be delivered; and based on the evaluation result of the material compatibility, the target material is determined from multiple materials to be delivered for delivery.

[0014] In an exemplary embodiment of the present disclosure, acquiring a target material segment from the material library based on the updated material effect score to generate a target material and delivering it includes: acquiring a material generation template; and based on the material generation template, generating multiple target materials from the target material segment and delivering them.

[0015] In an exemplary embodiment of the present disclosure, the collecting of user feedback features of the delivered material includes: obtaining a target user category for the delivery of the material; and obtaining user feedback features of the delivered material corresponding to the target user category according to the target user category.

[0016] According to one aspect of the present disclosure, a material delivery device is provided, including: a feature processing module, used to collect user feedback features of delivered materials, and determine first reward data of the delivered materials based on the user feedback features, wherein the first reward data reflects the delivery effect of the delivered materials, and the delivered materials are generated based on at least one material fragment obtained from a material library; a reward attribution module, used to perform material attribution processing based on the first reward data to determine second reward data of the material fragment; and a material generation module, used to update a material effect score of the material fragment based on the second reward data and a historical material indicator of the material fragment, and acquire a target material fragment from the material library based on the updated material effect score to generate a target material and deliver the target material.

[0017] According to one aspect of the present disclosure, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, any one of the above methods is implemented.

[0018] According to one aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform any one of the above methods by executing the executable instructions.

[0019] The material delivery method in the exemplary embodiment of the present disclosure collects user feedback features of delivered materials, and determines the first reward data of the delivered materials based on the user feedback features, the first reward data reflects the playback effect of the delivered materials, and the delivered materials are generated based on the material clips obtained from the material library; performs material attribution processing according to the first reward data to determine the second reward data of the material clips; updates the material effect score of the material clips according to the second reward data and the historical material indicators of the material clips, and obtains the target material clips from the material library based on the updated material effect score to generate the target material and deliver it. On the one hand, the material delivery effect can be evaluated by collecting user feedback features of the delivered materials, and the material delivery effect can be attributed to the material clips, and then the material effect score of the material clips in the material library can be updated to control the subsequent material selection and generate the delivered materials. This process drives the selection of the best materials based on the user feedback features, improves the quality of the delivered materials, makes them meet the actual interests and needs of different user groups, and improves the efficiency of material delivery. On the other hand, this process can continuously optimize the quality of the generated target material fragments based on user feedback through a cyclic execution process of generating materials, delivering materials, collecting information, evaluating materials, and optimizing materials, thereby reducing the complexity and time cost of manual intervention and improving the efficiency of material delivery.

[0020] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become readily understood by reading the following detailed description with reference to the accompanying drawings, in which several embodiments of the present disclosure are shown in an exemplary and non-limiting manner.

[0022] Figure 1 An application environment according to an exemplary embodiment of the present disclosure is shown.

[0023] Figure 2 A flow chart of a material delivery method according to an exemplary embodiment of the present disclosure is shown.

[0024] Figure 3 A flowchart of an implementation method for determining first reward data according to an exemplary embodiment of the present disclosure is shown.

[0025] Figure 4 A flowchart of an implementation method for determining second reward data according to an exemplary embodiment of the present disclosure is shown.

[0026] Figure 5 A flowchart is shown of an implementation method for updating a material effect score of a material clip according to an exemplary embodiment of the present disclosure.

[0027] Figure 6 A flow chart of a material selection method based on a policy according to an exemplary embodiment of the present disclosure is shown.

[0028] Figure 7 A schematic diagram showing the composition of a material delivery device according to an exemplary embodiment of the present disclosure is shown.

[0029] Figure 8 A block diagram of an electronic device according to an exemplary embodiment of the present disclosure is shown.

[0030] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0031] The exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, the exemplary embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the present disclosure will be more comprehensive and complete and the concepts of the exemplary embodiments will be fully conveyed to those skilled in the art. The same reference numerals in the figures represent the same or similar structures, and thus their detailed description will be omitted.

[0032] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known structures, methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the present disclosure.

[0033] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or these functional entities or parts of functional entities may be implemented in one or more software hardened modules, or these functional entities may be implemented in different networks and / or processor devices and / or microcontroller devices.

[0034] Currently, when preparing the materials for delivery, decisions are made manually on material selection, editing optimization, and effect evaluation. However, manually generating delivery materials is inefficient, cannot accurately capture user interests and market demand, and cannot know the delivery effect of the materials before delivery, resulting in low delivery efficiency and poor delivery effect.

[0035] Based on this, the exemplary embodiments of the present disclosure provide a material delivery method, which improves the efficiency and quality of delivered materials by collecting user feedback features of delivered materials and using them to control the selection of material segments. It should be noted that the material delivery method of the exemplary embodiments of the present disclosure can be applied to various material delivery scenarios, such as delivering videos, audios, pictures, texts, or various combinations of material segments (such as audio + video) on applications or web pages. The exemplary embodiments of the present disclosure do not specifically limit the application fields of the material delivery method. The exemplary embodiments of the present disclosure are described by taking the delivery target material as a video as an example.

[0036] The material delivery method provided by the exemplary embodiment of the present disclosure can be applied to Figure 1 In the application environment shown, the terminal 101 communicates with the server 102 via a network. The data storage system can store data that the server 102 needs to process. The data storage system can be integrated on the server 102, or placed on the cloud or other network servers.

[0037] In an exemplary embodiment, the material delivery method provided by the exemplary embodiment of the present disclosure may be executed by the server 102, and the corresponding material delivery device is disposed in the server 102. Correspondingly, in this mode of execution by the server 102, the server 102 may start executing the steps in the technical solution of the exemplary embodiment of the present disclosure in response to a trigger command, wherein the trigger command may be sent by a terminal used by a user, or may be triggered locally by the server in response to some automation events.

[0038] The server 102 may be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server 102 may perform background tasks.

[0039] Furthermore, in another exemplary embodiment, the terminal 101 may also have similar functions to the server 102 , so as to execute the material delivery method provided by the exemplary embodiment of the present disclosure.

[0040] The terminal 101 may be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, an IoT device, and a portable wearable device. The IoT device may be a smart speaker, a smart TV, a smart air conditioner, and a smart vehicle-mounted device, etc. The portable wearable device may be a smart watch, a smart bracelet, a head-mounted device, etc. The terminal 101 may also be referred to as a mobile terminal, a terminal device, a mobile device, etc. The exemplary embodiments of the present disclosure do not limit the type of the terminal 101.

[0041] In addition, the technical solution of the exemplary embodiment of the present disclosure can also be executed collaboratively by the terminal 101 and the server 102. In this manner of collaborative execution by the terminal 101 and the server 102, some steps in the technical solution provided by the exemplary embodiment of the present disclosure are executed by the terminal 101, while other steps are executed by the server 102. It should be noted that in this manner of collaborative execution by the terminal 101 and the server 102, the steps respectively executed by the terminal 101 and the server 102 can be dynamically adjusted according to actual conditions, and there is no special limitation on this.

[0042] The terminal 101 and the server 102 may be connected directly or indirectly via wireless communication, and the exemplary embodiments of the present disclosure do not impose any special limitation thereto.

[0043] It should be noted that the information involved in the embodiments of the present disclosure (including but not limited to: information input by the user, for example, information input by the user into the input box), data (including but not limited to data used for analysis, stored data, displayed data, for example, user feedback characteristics of the released materials, historical material indicators of the material clips, and user categories, etc.) are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards. For example, the user feedback characteristics of the released materials, historical material indicators of the material clips, and user categories, etc., involved in the embodiments of the present disclosure are all obtained with full authorization.

[0044] In an exemplary embodiment of the present disclosure, a material delivery method is provided, referring to Figure 2 The flowchart of the material delivery method of the exemplary embodiment of the present disclosure is shown, and the material delivery method includes steps S210 to S230, which are specifically as follows:

[0045] In step S210, user feedback features of the delivered material are collected, and first reward data of the delivered material is determined based on the user feedback features. The first reward data reflects the delivery effect of the delivered material, and the delivered material is generated based on material clips obtained from the material library.

[0046] In the exemplary embodiment of the present disclosure, the delivered material refers to the material that has been delivered on the target platform, and the target platform can be a platform such as a social media platform that can deliver the material. By delivering the material to the target platform, the user feedback characteristics of the delivered material can be obtained. The user feedback characteristics refer to indicators that reflect the performance of the material, including exposure, playback, completion rate, bounce rate, interaction rate (such as the number of interactive behaviors such as the number of likes, comments, and forwarding rates), viewing time, number of member views, etc. Based on the user feedback characteristics, the user's emotional response and interactive behavior to the played material can be known. Among them, exposure refers to the number of times the delivered material is displayed, playback refers to the number of times the delivered material is clicked to play, completion rate refers to the proportion of users who have watched the delivered material completely, and bounce rate refers to the proportion of users who leave during the process of watching the material (such as watching a video). The exemplary embodiment of the present disclosure can collect the above-mentioned user feedback characteristics in real time by deeply integrating the data collection module with the interface of the target platform, and there is no restriction on this.

[0047] Among them, the material library is used to store material clips. Each time the material is generated, the material clips are obtained from the material library. The quality of the selected material clips affects the playback effect of the generated material to a certain extent. The first reward data can reflect the performance effect of the delivery of the delivered material. The larger the first reward data, the better the playback effect of the delivered material. The first reward data can be determined by combining multiple user feedback features of the delivered material, so that the first reward data can comprehensively reflect the comprehensive performance of the delivered material in multiple feature dimensions. It should be understood that a piece of audio can be used as a material clip, a piece of video (without audio) can also be used as a material clip, a video (with audio) can also be used as a material clip, and even a piece of text and a picture can be used as a material clip, and there is no special limitation on this. Exemplarily, a video can be generated by selecting a piece of music, some pictures and a piece of text as the target material for delivery. The embodiment of the present disclosure does not impose special restrictions on the type of target material and the specific type of material clips contained in the target material.

[0048] It should be understood that "play" for audio and video means broadcasting audio or video through radio or television (or via the Internet) for users to watch or listen to. For pictures or text, it may mean providing a form for browsing pictures or texts. This will not be repeated later.

[0049] In step S220, material attribution processing is performed according to the first reward data to determine second reward data of the material segment.

[0050] In an exemplary embodiment of the present disclosure, material attribution processing refers to allocating first reward data corresponding to the delivered material to each material segment to obtain second reward data of the material segment, that is, the contribution degree of the material segment to the performance of the delivered material.

[0051] Among them, considering that users may jump out at any time during the process of watching materials, the closer the material clips are displayed to the front, the greater the contribution to the material performance, so the material attribution processing can be performed according to the position of the material clips in the clips of the played materials.

[0052] In step S230, the material effect score of the material clip is updated according to the first reward data and the historical material index of the material clip, and the target material clip is obtained from the material library based on the updated material effect score to generate the target material and release it.

[0053] In an exemplary embodiment of the present disclosure, the historical material index of a material clip refers to performance information reflecting the material clip when generating historical materials, including at least one of the number of times the material clip is selected, the number of times the material is played, and the number of times the material is interacted. Among them, the number of times selected indicates the number of times the material clip is used to generate historically delivered materials, the number of times the material is played completely, and the number of times the material is interacted indicates the total number of times the historically delivered materials are interacted during the playback of the material clip.

[0054] The material effect score of a material clip is a score that reflects the material playback effect. A higher material effect score indicates that the user has a stronger interest in the content of the material clip. Correspondingly, a higher material effect score increases the probability that the material clip will be selected when generating materials in the future.

[0055] For example, after obtaining the second reward data and historical material index of each material that has been released, the material effect score of the material clip can be updated to control the material selection when generating a video next time. When generating a video by obtaining the target material clip according to the updated material effect score, the number of videos generated each time can be flexibly set according to actual needs, and there is no restriction on this.

[0056] It should be noted that the exemplary embodiment of the present disclosure repeatedly executes the above steps S210 to S230, and continuously updates the material effect score of the material clip through user feedback on the released material, so as to optimize the quality of the material clip for generating subsequent materials and improve the quality and efficiency of material release.

[0057] The material delivery method in the exemplary embodiment of the present disclosure, on the one hand, can evaluate the material delivery effect by collecting user feedback features of the delivered materials, and attribute the material delivery effect to the material segments, and then can update the material effect scores of the material segments in the material library, so as to control the subsequent material selection and generate delivery materials. This process drives the selection of materials based on user feedback features, improves the quality of delivered materials, makes them meet the actual interests and needs of users, and improves the efficiency of material delivery. On the other hand, this process can continuously optimize the quality of the material segments of the generated target materials based on user feedback through the cyclic execution process of generating materials, delivering materials, collecting information, evaluating materials, and optimizing materials, thereby reducing the complexity and time cost of manual intervention and improving the efficiency of material delivery.

[0058] In an exemplary embodiment, a method for determining first reward data is provided. Figure 3 As shown, determining the first reward data of the delivered material based on the user feedback feature may include:

[0059] Step S310: Obtain weight information corresponding to the user feedback feature, where the weight information reflects the importance of the user feedback feature in the first reward data. The weight information corresponding to the user feedback feature can be set according to the actual material delivery task, and there is no specific limitation on this.

[0060] Step S320: Based on the weight information, the user feedback features corresponding to the delivered materials are integrated to obtain the first reward data.

[0061] The collected user feedback features may be preprocessed, including but not limited to data cleaning, standardization, normalization, and removal of outliers, to form high-quality indicator data for analysis. Among them, the user feedback features corresponding to the delivered materials may be weighted and fused based on the weight information to obtain the first reward data.

[0062] For example, the released material is a video, and the user feedback features include the playback volume, completion rate, and interaction rate. The first reward data can be obtained by the following formula:

[0063] R t =α×(V t / V max )+β×C t +γ×E t (1)

[0064] Among them, α, β and γ are the weight information corresponding to each user feedback feature, V t is the number of views of the released video (taking the tth released video as an example), C t is the completion rate of the delivered videos, E tis the interaction rate, V max It is the maximum number of views among all released videos, which is used for data normalization. Of course, other indicator data can also be selected for data normalization based on the user feedback features used to calculate the first reward data.

[0065] It should be noted that the first reward data can be obtained by using formula (1) for other types of released materials, which will not be described in detail here.

[0066] By fusing the user feedback features corresponding to the released materials to obtain the first reward data, the performance of the released materials can be comprehensively evaluated, providing an accurate data basis for the subsequent determination of the second reward data.

[0067] In an exemplary embodiment, a method for determining the second reward data is provided. Figure 4 As shown, performing material attribution processing according to the first reward data to determine the second reward data of the material segment may include:

[0068] Step S410: determining a material weight corresponding to the material segment according to the segment position of the material segment in the delivered material.

[0069] Considering that users may jump out at any time while watching the material, the material clips that are closer to the front in the material should be assigned higher material weights. Material weights can be assigned to material clips based on a weighted delivery model associated with the average playback time of users, wherein a weighted decreasing model can be pre-established to characterize the change of material weights with the position of the clips, thereby determining the material weights corresponding to the material clips according to the position of the clips based on the preset weighted decreasing model. Among them, the sum of the material weights of the material clips that have been put into the material is 1.

[0070] For example, if the delivered video contains n material clips, and the average user playback time covers the first m material clips, the material weight of the i-th material clip can be determined by the following linear decreasing model:

[0071] w i1 =w0-k×i (2)

[0072] Among them, w0 is the initial weight, k is the decreasing coefficient, both of which can be set according to needs, and i is the position of the material clip in the released video.

[0073] Exemplarily, the material weight of the i-th material segment may be determined by the following exponential decrease model:

[0074] w i2 =w0×exp(-λ×i) (3)

[0075] Among them, λ is the decreasing rate, which can be set according to needs.

[0076] Step S420: Determine second reward data for the material segment based on the material weight and the first reward data.

[0077] After obtaining the material weight and the first reward data, the second reward data of the material segment can be obtained through the material weight and the first reward data, that is, the first reward data can be multiplied by the material weight of the material segment to obtain the second reward data of the material segment.

[0078] Among them, the linear decreasing model can constrain the material weight of each material clip to gradually decrease at a fixed amplitude, while the exponential decreasing model constrains the material weight to decrease exponentially with the material position, so that the importance of different material clips can be flexibly adjusted. Through the design of the weight decreasing model, the material clips that are closer to the front can obtain higher weights, because even in the case of a high bounce rate, the material clips at the front have a greater impact on the user's overall visual experience. By attributing the first reward data to each material clip, it is possible to clarify the contribution of each material clip to the performance of the played material, thereby ensuring that the content seen by the user at the front of the material has a greater impact on the overall effect, that is, by reasonably allocating the material weights of the material clips, it is beneficial to improve the overall effect of the material.

[0079] It should be noted that, for other types of released materials, the material weight of the i-th material segment can be obtained by using formulas (2) and (3), which will not be described in detail here.

[0080] In an exemplary embodiment, a method for updating the material effect score of a material clip is also provided. Figure 5 As shown, updating the material effect score of the material clip according to the second reward data and the historical material index of the material clip may include:

[0081] Step S510: Obtain the current material reward corresponding to the material clip.

[0082] The current material reward refers to the material reward to be updated of the material segment, that is, the material reward obtained after the last round of update. For example, it can be set to update the material reward once every preset period, and the current material reward is the material reward (second reward data) obtained in the last period update. The exemplary embodiment of the present disclosure does not specifically limit the preset period.

[0083] Step S520: Obtain the reward difference value between the current material reward and the second reward data, wherein the difference between the current material reward and the second reward data can be obtained as the reward difference value.

[0084] Step S530: determining a reward correction parameter based on the historical material index, and updating the material effect score of the material clip according to the reward correction parameter, the reward difference value and the current material reward.

[0085] As can be seen from the above, the historical material index may include at least one of the number of times a material segment is selected, the number of times a material is finished playing, and the number of times a material is interacted. Taking the case where the historical material index includes the number of times a material segment is selected as an example, the reciprocal of the number of times selected can be obtained as a reward correction parameter, and if the historical material index includes the number of times a material segment is selected, the number of times a material is finished playing, and the number of times a material is interacted, then the sum of the number of times selected, the number of times a material is finished playing, and the number of times a material is interacted can be obtained, and then the sum can be taken as a reward correction parameter. The exemplary embodiment of the present disclosure determines the reward correction parameter by using the historical material index, so that when the material effect score is updated based on the material reward (current material reward, second reward data), the playback performance of the material segment itself can be fully considered, and the accuracy of the updated material effect score can be further improved.

[0086] According to the reward correction parameter, the reward difference value and the current material reward, the material effect score of the material clip is updated. The reward difference value can be corrected (such as multiplied) by the reward correction parameter, and then the correction result is combined with the current material reward to obtain the updated material effect score. Exemplarily, the material effect score of the material clip can be updated by the following formula:

[0087] Q i' =Q i +(1 / N i )×(R i -Q i ) (4)

[0088] Among them, Q i is the current material reward of the material clip, Q i’ Score the updated material effect of the material clip, 1 / N i is the reward correction parameter, R i is the second reward data, (R i -Q i ) is the reward difference value.

[0089] By updating the material effect scores of the material clips as described above, the performance of the material clips can be gradually adjusted, so that material clips with high user feedback can obtain higher performance effect scores, thereby increasing the probability of being used to generate target materials in the future, and also improving the quality of materials used to generate delivery materials, thereby improving the quality of delivery materials.

[0090] As an example, if the current material reward of a material clip i is Q i= 0.5, and the number of times the clip is selected after delivery is N i =1, the second reward data obtained by calculation is 0.8, then by substituting into formula (4), the updated material effect score is: Q i’ =0.5+(1 / 1)×(0.8-0.5)=0.8, that is, the material effect score of the material clip is increased from 0.5 to 0.8, indicating that the material clip performs better in this delivery.

[0091] By adjusting the material effect score of the material clip based on the performance of the material clip in the released material, the material clip with a higher material effect score is more likely to be used in the subsequent material generation, thereby improving the release effect of the material. For example, if the average viewing time of the released video containing a material clip of natural landscape is higher than that of other videos, and the material effect score of the material clip is improved, then in the next round of material selection, the material clip containing natural landscape will be more likely to be selected based on the updated material effect score, or it will be combined with other materials (such as new music) to generate a new video version.

[0092] By gradually optimizing the selection of materials through multiple iterations, we can eventually obtain the most effective material combination and continuously improve the overall material delivery effect.

[0093] In an exemplary embodiment, a strategy-based material selection method is also provided, such as Figure 6 As shown, obtaining a target material clip from a material library based on the updated material effect score to generate a target material and delivering the target material includes:

[0094] Step S610: Obtaining a material exploration probability, where the material exploration probability is used to indicate the proportion of random materials in the target material segment.

[0095] The material exploration probability is the material selection probability based on the greedy strategy, which means that when generating a target, some materials will be randomly selected from the material library according to a certain ratio to ensure the exploration of the material fragments in the entire material library and prevent falling into the local optimum. Specifically, the material exploration probability is pre-set, that is, the proportion of random materials in each target material, for example, the material exploration probability ∈=0.1.

[0096] Step S620: obtaining random materials from the material library according to the material exploration probability, and obtaining generated materials from the material library based on the updated material effect score, so as to determine the target material segment according to the random materials and the generated materials and generate the target material for delivery.

[0097] You can first obtain material probability from the material library based on the material exploration probability, rather than selecting based on the material effect score of the material clip, and then obtain generated material from the material library based on the final material effect score, so as to generate target material based on the target material clip determined by the random material and the generated material for delivery.

[0098] Among them, when obtaining generated materials from the material library based on the updated material effect score, the material clips in the material library can be sorted in descending order according to the material effect score, and a preset number of material clips with the highest scores can be obtained as generated material clips. The preset number can be set according to actual needs to ensure the quality of the material clips.

[0099] Through the material exploration strategy, new material combinations are tried with a certain probability. Even if the material effect score of the material clips in the material library is low, the material will be randomly selected according to the material exploration probability to ensure the exploration of the entire material library, so as to tap into potential excellent materials, especially those that are not fully utilized. The exploration and utilization can be balanced by constantly adjusting the size of the material exploration probability, so as to adapt to the dynamic changes of the material library at different stages, ensuring that both potential excellent material clips can be fully tapped and currently known high-quality material clips can be utilized to improve the overall material delivery effect.

[0100] In an exemplary embodiment, a method for optimizing the material delivery effect is also provided. The material clips in the material library have their own corresponding category labels, such as real-life shooting, ancient style, high-level, lyrical, CG (Computer Graphics), etc., which can be set according to needs. The compatibility of material clips of the same type is higher, for example, real-life videos and real-life videos are more compatible, while real-life videos and CG animations are less compatible.

[0101] Among them, first, based on the updated material effect score, the target material fragments can be obtained from the material library to generate multiple materials to be released; for each material to be released, the material compatibility of the material to be released is evaluated according to the category label of the target material fragment contained in the material to be released; finally, according to the evaluation result of the material compatibility, the target material is determined from the multiple materials to be released for release.

[0102] Exemplarily, for each material to be delivered, the correlation between the category labels of the target material segments included in the material to be delivered can be determined. Optionally, an evaluation model for determining material compatibility can be pre-trained to determine the compatibility of the material segments in each material to be delivered. Optionally, association data between different material categories can also be pre-set, and the association data can be queried according to the category labels of the material segments in the material to be delivered, so as to determine the material compatibility of the material to be delivered according to the query results.

[0103] By further judging the material compatibility of the material to be delivered, the material compatibility of the target material can be further optimized and the quality of the delivered material can be improved.

[0104] In an exemplary embodiment, acquiring a target material segment from a material library based on an updated material effect score to generate a target video material and delivering the target video material comprises:

[0105] A material generation template is obtained, and based on the material generation template, a target material segment is generated into multiple target materials and then delivered.

[0106] The material generation template is the guiding information for controlling the material generation, including but not limited to the video frame structure, video editing method, editing point and transition method, text editing method, and the combination method of different types of material clips, etc. That is to say, when the exemplary embodiment of the present disclosure generates multiple materials to be delivered according to the target material clip, the same material generation template is used, and only the selected material clips are diversified to form multiple different versions of the target material. Based on this, the impact of the material clip on the performance of the delivered material can be more accurately quantified and evaluated, avoiding the impact of other variables on the evaluation of the material clip, and improving the accuracy of the subsequent selection of the material clip.

[0107] The material generation template may be provided manually or randomly generated by the system, and there is no specific limitation on this.

[0108] In an exemplary embodiment, collecting user feedback features of delivered materials may include:

[0109] The target user category for material delivery is obtained, and based on the target user category, user feedback features of the delivered material corresponding to the target user category are obtained.

[0110] Among them, user categories can be set according to age groups or other user characteristics, such as gender, occupation, etc., and can be flexibly set according to actual needs.

[0111] Considering that different clips have different appeals to different groups of users, the data range relied on for calculating the material effect score can be adjusted according to different user categories by selecting the user categories for material delivery. For example, when calculating the first reward data, the calculation can be based on values ​​such as the play rate / interaction rate / complete rate. If this part of the data range is set within the young people range, the first reward data is aimed at rewards that are more popular with young people. That is, without changing the formula itself, by adjusting the specific value source, the first reward data can be adapted to the needs of different delivery population characteristics, and then select clips that are more popular with young people to generate target materials for delivery.

[0112] The material delivery method in the exemplary embodiment of the present disclosure collects user feedback features of delivered materials, and determines the first reward data of the delivered materials based on the user feedback features, the first reward data reflects the delivery effect of the delivered materials, and the delivered materials are generated based on the material clips obtained from the material library; performs material attribution processing according to the first reward data to determine the second reward data of the material clips; updates the material effect score of the material clips according to the second reward data and the historical material indicators of the material clips, and obtains the target material clips from the material library based on the updated material effect score to generate the target material and deliver it. On the one hand, the material playback effect can be evaluated by collecting user feedback features of the delivered materials, and the material playback effect can be attributed to the material clips, and then the material effect score of the material clips in the material library can be updated to control the subsequent material selection and generate the delivery material. This process drives the selection of the best materials based on the user feedback features, improves the quality of the delivered materials, makes them meet the actual interests and needs of different user groups, and improves the efficiency of material delivery. On the other hand, this process can continuously optimize the quality of generated materials based on user feedback through a cyclic execution process of generating materials, delivering materials, collecting information, evaluating materials, and optimizing materials, thereby reducing the complexity and time cost of manual intervention and improving the efficiency of material delivery.

[0113] In an exemplary embodiment of the present disclosure, a material delivery device is also provided. Figure 7 As shown, the material delivery device 700 may include a feature processing module 710, a reward attribution module 720 and a material generation module 730. Specifically:

[0114] The feature processing module 710 is used to collect user feedback features of the released material, and determine the first reward data of the released material based on the user feedback features, wherein the first reward data reflects the release effect of the released material, and the released material is generated based on the material fragment obtained from the material library; the reward attribution module 720 is used to perform material attribution processing based on the first reward data to determine the second reward data of the material fragment; the material generation module 730 is used to update the material effect score of the material fragment based on the second reward data and the historical material index of the material fragment, and obtain the target material fragment from the material library based on the updated material effect score to generate the target material and release it.

[0115] In an exemplary embodiment of the present disclosure, the feature processing module 710 is configured to perform: obtaining weight information corresponding to the user feedback feature, the weight information reflecting the importance of the user feedback feature in the first reward data; based on the weight information, fusing the user feedback features corresponding to the delivered materials to obtain the first reward data.

[0116] In an exemplary embodiment of the present disclosure, the reward attribution module 720 is configured to execute: determining a material weight corresponding to the material clip according to the clip position of the material clip in the released material; and determining second reward data for the material clip based on the material weight and the first reward data.

[0117] In an exemplary embodiment of the present disclosure, the reward attribution module 720 is configured to execute: based on a preset weight reduction model, determine the material weight corresponding to the material segment according to the segment position; wherein the weight reduction model is used to characterize the change of the material weight with the segment position.

[0118] In an exemplary embodiment of the present disclosure, the material generation module 730 is configured to execute: obtaining a current material reward corresponding to the material clip; obtaining a reward difference value between the current material reward and the second reward data; determining a reward correction parameter based on the historical material indicator, and updating the material effect score of the material clip according to the reward correction parameter, the reward difference value and the current material reward.

[0119] In an exemplary embodiment of the present disclosure, the historical material indicator includes at least one of the number of times the material segment is selected, the number of times the material is completed, and the number of times the material is interacted; wherein the number of times the material segment is selected indicates the number of times the material segment is used to generate historically released materials, the number of times the material is completed indicates the number of times the historically released materials are completely played, and the number of material interactions indicates the total number of interactions that occurred with the historically released materials during the playback of the material segment.

[0120] In an exemplary embodiment of the present disclosure, the material generation module 730 is configured to perform: obtaining a material exploration probability, wherein the material exploration probability is used to indicate the proportion of random materials in the target material segment; obtaining random materials from the material library according to the material exploration probability, and obtaining generated materials from the material library based on the updated material effect score, so as to determine the target material segment based on the random materials and the generated materials and generate target materials for delivery.

[0121] In an exemplary embodiment of the present disclosure, the material clips in the material library have respective corresponding category labels; the material generation module 730 is configured to execute: based on the updated material effect score, obtaining the target material clip from the material library to generate multiple materials to be delivered; for each material to be delivered, evaluating the material compatibility of the material to be delivered according to the category label of the target material clip contained in the material to be delivered; and determining the target material from the multiple materials to be delivered for delivery according to the evaluation result of the material compatibility.

[0122] In an exemplary embodiment of the present disclosure, the material generation module 730 is configured to execute: obtaining a material generation template; and based on the material generation template, generating a plurality of target materials from the target material segment and delivering the generated materials.

[0123] In an exemplary embodiment of the present disclosure, the feature processing module 710 is configured to execute: obtaining a target user category for material delivery; and obtaining user feedback features of the delivered material corresponding to the target user category according to the target user category.

[0124] Since the details of the various functional modules of the material delivery device of the exemplary embodiment of the present disclosure have been recorded in the exemplary embodiment of the material delivery method described above, they will not be repeated here.

[0125] It should be noted that although several modules or units of the material delivery device are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units for embodiment.

[0126] The exemplary embodiment of the present disclosure further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the above-mentioned material delivery method is implemented.

[0127] In one embodiment, the computer program product may be a tangible product containing a computer program, such as a computer-readable storage medium storing a computer program. The readable storage medium may be a storage medium based on electrical, magnetic, optical, electromagnetic, infrared, or other signals, including but not limited to: random access memory (RAM), read-only memory (ROM), magnetic tape, floppy disk, flash memory (Flash), mechanical hard disk (HDD), solid-state drive (SSD), and the like. Exemplarily, the computer program product may be implemented as a non-volatile storage medium storing a computer program, such as a read-only memory, a NAND flash memory, and the like.

[0128] In one embodiment, the computer program product may be an intangible product including a computer program. Exemplarily, the computer program product may be implemented as a virtual digital product, such as a digital file storing an executable file, an installation package, etc. of the computer program.

[0129] The code of the computer program can be written in one or more programming languages. Programming languages ​​such as C language, Java, C++, etc. The program code can be executed entirely on the user computing device, or partially on the user computing device, or as a separate software package, or partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, such as a local area network (LAN), a wide area network (WAN), etc., or can be connected to an external computing device (e.g., an Internet connection provided by an operator).

[0130] The computer program can be carried or transmitted through electrical, magnetic, optical, electromagnetic, infrared and other signals. The electronic device can convert the signal carrying the computer program into a digital signal, and then run the computer program. When the computer program runs on the electronic device, its code is used to make the electronic device execute (more specifically, the processor of the electronic device can execute) the method steps of various exemplary embodiments of the present disclosure, such as the above-mentioned material delivery method, which includes the following steps: collecting user feedback features of the delivered material, and determining the first reward data of the delivered material based on the user feedback features, the first reward data reflects the delivery effect of the delivered material, and the delivered material is generated based on the material fragment obtained from the material library; performing material attribution processing according to the first reward data to determine the second reward data of the material fragment; updating the material effect score of the material fragment according to the second reward data and the historical material index of the material fragment, and acquiring the target material fragment from the material library based on the updated material effect score to generate the target material and deliver it.

[0131] By executing the above method steps through a computer program, on the one hand, the material playback effect can be evaluated by collecting user feedback features of the released materials, and the material playback effect can be attributed to the material clips, and then the material effect score of the material clips in the material library can be updated to control the subsequent material selection and generate the released materials. This process drives the selection of materials based on user feedback features, improves the quality of the released materials, makes them meet the actual interests and needs of different user groups, and improves the efficiency of material delivery. On the other hand, this process can continuously optimize the material quality of the generated materials based on user feedback through the cyclic execution process of generating materials, releasing materials, collecting information, evaluating materials, and optimizing materials, reducing the complexity and time cost of manual intervention, and improving the efficiency of material delivery.

[0132] In an exemplary embodiment of the present disclosure, determining the first reward data of the released material based on the user feedback features includes: obtaining weight information corresponding to the user feedback features, the weight information reflecting the importance of the user feedback features in the first reward data; based on the weight information, fusing the user feedback features corresponding to the released material to obtain the first reward data.

[0133] In an exemplary embodiment of the present disclosure, the material attribution processing is performed according to the first reward data to determine the second reward data of the material fragment, including: determining the material weight corresponding to the material fragment according to the fragment position of the material fragment in the released material; and determining the second reward data of the material fragment based on the material weight and the first reward data.

[0134] In an exemplary embodiment of the present disclosure, determining the material weight corresponding to the material fragment according to the fragment position of the material fragment in the already-released material includes: determining the material weight corresponding to the material fragment according to the fragment position based on a preset weight decreasing model; wherein the weight decreasing model is used to characterize the change of the material weight with the fragment position.

[0135] In an exemplary embodiment of the present disclosure, updating the material effect score of the material clip according to the second reward data and the historical material indicators of the material clip includes: obtaining a current material reward corresponding to the material clip; obtaining a reward difference value between the current material reward and the first reward data; determining a reward correction parameter based on the historical material indicators, and updating the material effect score of the material clip according to the reward correction parameter, the reward difference value and the current material reward.

[0136] In an exemplary embodiment of the present disclosure, the historical material indicator includes at least one of the number of times the material segment is selected, the number of times the material is completed, and the number of times the material is interacted; wherein the number of times the material segment is selected indicates the number of times the material segment is used to generate historically released materials, the number of times the material is completed indicates the number of times the historically released materials are completely played, and the number of material interactions indicates the total number of interactions that occurred with the historically released materials during the playback of the material segment.

[0137] In an exemplary embodiment of the present disclosure, the target material fragment is obtained from the material library based on the updated material effect score to generate the target material and deliver it, including: obtaining the material exploration probability, the material exploration probability is used to indicate the proportion of random material in the target material fragment; obtaining random material from the material library according to the material exploration probability, and obtaining generated material from the material library based on the updated material effect score, so as to determine the target material fragment according to the random material and the generated material and generate the target material for delivery.

[0138] In an exemplary embodiment of the present disclosure, material clips in a material library have respective corresponding category labels; based on updated material effect scores, target material clips are obtained from the material library to generate a video and then delivered, including: based on the updated material effect scores, target material clips are obtained from the material library to generate multiple materials to be delivered; for each material to be delivered, the material compatibility of the material to be delivered is evaluated according to the category label of the target material clip included in the material to be delivered; and based on the evaluation result of the material compatibility, the target material is determined from the multiple materials to be delivered for delivery.

[0139] In an exemplary embodiment of the present disclosure, acquiring a target material segment from the material library based on the updated material effect score to generate a target material and delivering it includes: acquiring a material generation template; and based on the material generation template, generating multiple target materials from the target material segment and delivering them.

[0140] In an exemplary embodiment of the present disclosure, the collecting of user feedback features of the released video includes: obtaining a target user category for material release; and obtaining user feedback features of the released material corresponding to the target user category according to the user target category.

[0141] In addition, in an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided. It will be appreciated by those skilled in the art that various aspects of the present disclosure may be implemented as a system, method, or program product. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which may be collectively referred to herein as a "circuit", "module", or "system".

[0142] Refer to the following Figure 8 800 according to this embodiment of the present disclosure is described. Figure 8 The electronic device 800 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0143] like Figure 8As shown, the electronic device 800 is in the form of a general computing device. The components of the electronic device 800 may include, but are not limited to: the at least one processing unit 810, the at least one storage unit 820, a bus 830 connecting different system components (including the storage unit 820 and the processing unit 810), and a display unit 840.

[0144] The storage unit stores a program code, and the program code can be executed by the processing unit 810, so that the processing unit 810 executes the steps described in the above "Exemplary Method" section of this specification according to various exemplary embodiments of the present disclosure. For example, the processing unit 810 can execute the following steps: collect user feedback features of the released video, and determine the first reward data of the released material based on the user feedback features, the first reward data reflects the release effect of the released material, and the released material is generated based on at least one material clip obtained from the material library; perform material attribution processing according to the first reward data to determine the second reward data of the material clip; update the material effect score of the material clip according to the second reward data and the historical material index of the material clip, and obtain the target material clip from the material library based on the updated material effect score to generate the target material and release it.

[0145] In an exemplary embodiment of the present disclosure, determining the first reward data of the released material based on the user feedback features includes: obtaining weight information corresponding to the user feedback features, the weight information reflecting the importance of the user feedback features in the first reward data; based on the weight information, fusing the user feedback features corresponding to the released material to obtain the first reward data.

[0146] In an exemplary embodiment of the present disclosure, the material attribution processing is performed according to the first reward data to determine the second reward data of the material fragment, including: determining the material weight corresponding to the material fragment according to the fragment position of the material fragment in the released material; and determining the second reward data of the material fragment based on the material weight and the first reward data.

[0147] In an exemplary embodiment of the present disclosure, determining the material weight corresponding to the material fragment according to the fragment position of the material fragment in the already-released material includes: determining the material weight corresponding to the material fragment according to the fragment position based on a preset weight decreasing model; wherein the weight decreasing model is used to characterize the change of the material weight with the fragment position.

[0148] In an exemplary embodiment of the present disclosure, updating the material effect score of the material clip according to the second reward data and the historical material indicators of the material clip includes: obtaining a current material reward corresponding to the material clip; obtaining a reward difference value between the current material reward and the second reward data; determining a reward correction parameter based on the historical material indicators, and updating the material effect score of the material clip according to the reward correction parameter, the reward difference value and the current material reward.

[0149] In an exemplary embodiment of the present disclosure, the historical material indicator includes at least one of the number of times the material segment is selected, the number of times the material is completed, and the number of times the material is interacted; wherein the number of times the material segment is selected indicates the number of times the material segment is used to generate historically released materials, the number of times the material is completed indicates the number of times the historically released materials are completely played, and the number of material interactions indicates the total number of interactions that occurred with the historically released materials during the playback of the material segment.

[0150] In an exemplary embodiment of the present disclosure, the target material fragment is obtained from the material library based on the updated material effect score to generate the target material and deliver it, including: obtaining the material exploration probability, the material exploration probability is used to indicate the proportion of random material in the target material fragment; obtaining random material from the material library according to the material exploration probability, and obtaining generated material from the material library based on the updated material effect score, so as to determine the target material fragment according to the random material and the generated material and generate the target material for delivery.

[0151] In an exemplary embodiment of the present disclosure, material clips in a material library have respective corresponding category labels; based on updated material effect scores, target material clips are obtained from the material library to generate target materials and then delivered, including: based on updated material effect scores, target material clips are obtained from the material library to generate multiple materials to be delivered; for each material to be delivered, the material compatibility of the material to be delivered is evaluated according to the category label of the target material clip included in the material to be delivered; and based on the evaluation result of the material compatibility, the target material is determined from multiple materials to be delivered for delivery.

[0152] In an exemplary embodiment of the present disclosure, acquiring a target material segment from the material library based on the updated material effect score to generate a target material and delivering it includes: acquiring a material generation template; and based on the material generation template, generating multiple target materials from the target material segment and delivering them.

[0153] In an exemplary embodiment of the present disclosure, the collecting of user feedback features of the delivered material includes: obtaining a target user category for the delivery of the material; and obtaining user feedback features of the delivered material corresponding to the target user category according to the target user category.

[0154] The storage unit 820 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 821 and / or a cache memory unit 822 , and may further include a read-only memory unit (ROM) 823 .

[0155] The storage unit 820 may also include a program / utility 824 having a set (at least one) of program modules 825, such program modules 825 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0156] Bus 830 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0157] The electronic device 800 may also communicate with one or more external devices 900 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 800, and / or communicate with any device that enables the electronic device 800 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 850. Furthermore, the electronic device 800 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 860. As shown, the network adapter 860 communicates with other modules of the electronic device 800 via a bus 830. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0158] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiment of the present disclosure.

[0159] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0160] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. A material delivery method, characterized in that: include: Collecting user feedback features of the delivered material, and determining first reward data of the delivered material based on the user feedback features, wherein the first reward data reflects delivery effect of the delivered material, wherein the delivered material is generated based on at least one material clip obtained from a material library; Performing material attribution processing according to the first reward data to determine second reward data of the material clip; According to the second reward data and the historical material index of the material clip, the material effect score of the material clip is updated, and based on the updated material effect score, a target material clip is acquired from the material library to generate a target material and deliver it.

2. The method according to claim 1, characterized in that The determining the first reward data of the delivered material based on the user feedback feature includes: Obtaining weight information corresponding to the user feedback feature, where the weight information reflects the importance of the user feedback feature in the first reward data; Based on the weight information, user feedback features corresponding to the delivered materials are integrated to obtain the first reward data.

3. The method according to claim 1, characterized in that The performing material attribution processing according to the first reward data to determine the second reward data of the material segment includes: Determining a material weight corresponding to the material segment according to a segment position of the material segment in the delivered material; Based on the material weight and the first reward data, second reward data for the material segment is determined.

4. The method according to claim 3, characterized in that The determining, according to the position of the material segment in the segment of the delivered material, a material weight corresponding to the material segment, includes: Based on a preset weight reduction model, determining a material weight corresponding to the material segment according to the segment position; The weight reduction model is used to characterize the change of material weight with the position of the segment.

5. The method according to claim 1, characterized in that The updating of the material effect score of the material clip according to the second reward data and the historical material index of the material clip includes: Obtaining the current material reward corresponding to the material clip; Obtaining a reward difference value between the current material reward and the second reward data; A reward correction parameter is determined based on the historical material indicator, and the material effect score of the material clip is updated according to the reward correction parameter, the reward difference value and the current material reward.

6. The method according to claim 5, characterized in that The historical material index includes at least one of the number of times the material segment is selected, the number of times the material is finished playing, and the number of times the material is interacted; Among them, the number of selections indicates the number of times the material segment is used to generate historically released materials, the number of material completions indicates the number of times the historically released materials completely play the material segment, and the number of material interactions indicates the total number of interactions that occur during the playback of the historically released materials.

7. The method according to claim 1, characterized in that The step of acquiring a target material segment from the material library based on the updated material effect score to generate a target material and delivering the target material comprises: Acquire a material exploration probability, where the material exploration probability is used to indicate a proportion of random materials in the target material segment; Random materials are acquired from the material library according to the material exploration probability, and generated materials are acquired from the material library based on the updated material effect score, so as to determine the target material segment according to the random materials and the generated materials and generate target materials for delivery.

8. The method according to claim 1, characterized in that The material clips in the material library have respective corresponding category labels; the step of acquiring the target material clip from the material library based on the updated material effect score to generate a video and deliver the video comprises: Acquire target material clips from the material library based on the updated material effect scores to generate multiple materials to be delivered; For each of the materials to be delivered, evaluating the material compatibility of the materials to be delivered according to the category label of the target material segment included in the materials to be delivered; According to the evaluation result of the material compatibility, a target material is determined from the multiple materials to be delivered for delivery.

9. The method according to claim 1, characterized in that: The step of acquiring a target material segment from the material library based on the updated material effect score to generate a target material and delivering the target material comprises: Get the material generation template; Based on the material generation template, the target material segment is used to generate multiple target materials and then delivered.

10. The method according to any one of claims 1 to 9, characterized in that The collecting of user feedback features of the delivered materials includes: Get the target user category for material delivery; According to the target user category, a user feedback feature of the delivered material corresponding to the target user category is obtained.

11. A material delivery device, characterized in that: The device comprises: a feature processing module, configured to collect user feedback features of the delivered material, and determine first reward data of the delivered material based on the user feedback features, wherein the first reward data reflects the delivery effect of the delivered material, and the delivered material is generated based on at least one material clip obtained from the material library; a reward attribution module, configured to perform material attribution processing according to the first reward data to determine second reward data of the material clip; The material generation module is used to update the material effect score of the material clip according to the second reward data and the historical material index of the material clip, and obtain the target material clip from the material library based on the updated material effect score to generate the target material and release it.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

13. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to perform the method of any one of claims 1 to 10 by executing the executable instructions.

Images