Multimedia material scheduling method, and note scheduling method and device of social platform

Abstract

The embodiment of the invention provides a multimedia material scheduling method and a note scheduling method and device for a social platform, and the method comprises the steps: determining a target project, a delivery time range corresponding to the target project, and a target delivery amount, determining a plurality of delivery time periods corresponding to the delivery time range, the method comprises the following steps: obtaining a target delivery amount, determining delivery components of the target delivery amount dispersed to a plurality of delivery time periods, matching the delivery components corresponding to the plurality of delivery time periods with residual available delivery amounts of corresponding capacity units, the maximum available delivery amount of the capacity units being obtained based on traffic predicted by a traffic prediction model for the corresponding delivery time periods, and the maximum available delivery amount of the capacity units being obtained based on the traffic predicted by the traffic prediction model for the corresponding delivery time periods. Under the condition that the target project is determined to be successfully distributed at this time according to the matching result of the delivery components corresponding to the plurality of delivery time periods, the plurality of delivery components are distributed to the matched capacity units, and the multimedia materials of the delivery components are delivered to the promotion positions of the social platform in the delivery time periods, so that the delivery congestion of the multimedia materials is avoided.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of computer technology, and in particular to a multimedia material scheduling method, a note scheduling method and device for a social platform. Background Art

[0002] With the booming development of social platforms, more and more users are sharing or browsing notes through social platforms to obtain information or conduct transactions. Notes, as a kind of multimedia material, include but are not limited to text, images, audio, video and other content. On social platforms, high-quality notes can usually obtain a large amount of natural exposure traffic. However, in some cases, such as preheating the volume of key activities and providing traffic support to publishers, platform intervention may be required to release notes through some scheduling strategies to increase the exposure of notes.

[0003] At present, the delivery strategy of notes is relatively simple, which is to set the total delivery amount for a period of time and deliver it according to the constraints of the total delivery amount. However, this simple delivery strategy often leads to congestion of delivery channels, and even the problem of delivery failure, resulting in poor delivery effect. Summary of the invention

[0004] In view of this, the embodiments of this specification provide a multimedia material scheduling method. One or more embodiments of this specification also involve a note scheduling method for a social platform, a multimedia material scheduling device, a note scheduling device for a social platform, a computing device, a computer-readable storage medium, and a computer program product to avoid congestion in delivery channels, improve delivery completion, and achieve better delivery effects.

[0005] According to a first aspect of an embodiment of the present specification, a multimedia material scheduling method is provided, comprising: in response to receiving a multimedia material delivery request, determining a target project corresponding to the multimedia material delivery request and a delivery time range and a target delivery amount corresponding to the target project; determining a plurality of delivery time periods corresponding to the delivery time range according to a preset time period division rule, and determining delivery components of the target delivery amount dispersed to the plurality of delivery time periods; matching the delivery components corresponding to the plurality of delivery time periods with the remaining available delivery amounts of the corresponding capacity units to obtain a matching result, wherein one capacity unit corresponds to one delivery time period, the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, and the maximum available delivery amount is obtained based on the traffic predicted by the traffic prediction model for the corresponding delivery time period; in the case where it is determined that the target project is successfully allocated this time according to the matching results of the delivery components corresponding to the plurality of delivery time periods, the plurality of delivery components are allocated to the matching capacity units; and the multimedia materials of the delivery components are delivered to the promotion positions of the social platform within the delivery time period of the allocated capacity unit.

[0006] According to a second aspect of an embodiment of the present specification, a note scheduling method for a social platform is provided, comprising: in response to receiving a note delivery request, determining a target project corresponding to the note delivery request and a delivery time range and a target delivery amount corresponding to the target project; determining a plurality of delivery time periods corresponding to the delivery time range according to a preset time period division rule, and determining delivery components of the target delivery amount dispersed to the plurality of delivery time periods; matching the delivery components corresponding to the plurality of delivery time periods with the remaining available delivery amounts of the corresponding capacity units to obtain a matching result, wherein one capacity unit corresponds to one delivery time period, the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, and the maximum available delivery amount is obtained based on the traffic predicted by a traffic prediction model for the corresponding delivery time period; in the case where it is determined that the target project is successfully allocated this time according to the matching results of the delivery components corresponding to the plurality of delivery time periods, the plurality of delivery components are allocated to the matching capacity units; and the notes of the delivery components are delivered to the promotion position of the social platform within the delivery time period of the allocated capacity unit.

[0007] According to a third aspect of the embodiments of this specification, a multimedia material scheduling device is provided, including: a project determination module, configured to determine, in response to receiving a multimedia material delivery request, a target project corresponding to the multimedia material delivery request and a delivery time range and a target delivery amount corresponding to the target project; a component determination module, configured to determine, according to a preset time period division rule, a plurality of delivery time periods corresponding to the delivery time range, and determine the delivery components of the target delivery amount dispersed into the plurality of delivery time periods; a component matching module, configured to match the delivery components corresponding to the plurality of delivery time periods respectively with the remaining available delivery amounts of the corresponding capacity units, A matching result is obtained, wherein one of the capacity units corresponds to one of the delivery time periods, the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, and the maximum available delivery amount is obtained based on the traffic predicted by the traffic prediction model for the corresponding delivery time period; a component allocation module is configured to allocate the multiple delivery components to the matching capacity unit when it is determined that the target project is successfully allocated this time according to the matching results of the delivery components corresponding to the multiple delivery time periods; a component delivery module is configured to deliver the multimedia materials of the delivery components to the promotion position of the social platform within the delivery time period of the allocated capacity unit.

[0008] According to a fourth aspect of the embodiments of the present specification, a note scheduling device for a social platform is provided, including: a note delivery response module, configured to determine, in response to receiving a note delivery request, a target project corresponding to the note delivery request and a delivery time range and a target delivery amount corresponding to the target project; a note delivery dispersion module, configured to determine, according to a preset time period division rule, a plurality of delivery time periods corresponding to the delivery time range, and determine delivery components of the target delivery amount dispersed to the plurality of delivery time periods; a note component matching module, configured to match the delivery components corresponding to the plurality of delivery time periods respectively with the remaining available delivery amounts of the corresponding capacity units. , obtaining a matching result, wherein one of the capacity units corresponds to one of the delivery time periods, the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, and the maximum available delivery amount is obtained based on the traffic predicted by the traffic prediction model for the corresponding delivery time period; a note component allocation module is configured to allocate the multiple delivery components to the matching capacity unit when it is determined that the target project is successfully allocated this time according to the matching results of the delivery components corresponding to the multiple delivery time periods; a note delivery module is configured to deliver the notes of the delivery components to the promotion position of the social platform within the delivery time period of the allocated capacity unit.

[0009] According to the fifth aspect of the embodiments of this specification, a computing device is provided, including: a memory and a processor; the memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions, which, when executed by the processor, implement the steps of the method described in any embodiment of this specification.

[0010] According to the sixth aspect of the embodiments of this specification, a computer-readable storage medium is provided, which stores a computer program / instruction, and when the computer program / instruction is executed by a processor, the steps of the method described in any embodiment of this specification are implemented.

[0011] According to a seventh aspect of the embodiments of this specification, a computer program product is provided, comprising a computer program / instruction, which, when executed by a processor, implements the steps of the method described in any embodiment of this specification.

[0012] An embodiment of the present specification implements a multimedia material scheduling method. The method, in response to receiving a multimedia material delivery request, determines a target project corresponding to the multimedia material delivery request and a delivery time range and a target delivery amount corresponding to the target project, determines a plurality of delivery time periods corresponding to the delivery time range according to a preset time period division rule, and determines delivery components of the target delivery amount dispersed to the plurality of delivery time periods, matches the delivery components corresponding to the plurality of delivery time periods with the remaining available delivery amounts of the corresponding capacity units, and obtains a matching result, wherein the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, and the maximum available delivery amount of the capacity unit corresponding to a delivery time period is obtained based on the traffic predicted by a traffic prediction model, therefore, when it is determined that the target project is successfully allocated this time according to the matching results of the delivery components corresponding to the plurality of delivery time periods, the plurality of delivery components are allocated to the matching capacity units, and the multimedia materials of the delivery components are delivered to the promotion positions of the social platform within the delivery time period of the allocated capacity units.

[0013] It can be seen that since this method obtains the maximum available delivery volume of the capacity unit corresponding to the delivery period based on the traffic predicted by the traffic prediction model, the maximum available delivery volume of the capacity unit is made to correspond to the difference between the high and low peaks in different time periods, and the target delivery volume of the multimedia materials of the target project is dispersed according to multiple delivery periods, and the delivery components corresponding to the dispersed multiple delivery time periods are matched with the remaining available delivery volume of the capacity unit. When it is determined according to the matching result that the target project is successfully allocated this time, the delivery components are allocated to the matching capacity units, thereby ensuring that the delivery is carried out under the condition of successful allocation, effectively ensuring the completion of the delivery, and the dispersed delivery components of the multimedia materials are accurately adapted to the difference between the high and low peaks in different time periods, which can effectively avoid concentrated delivery, avoid congestion in delivery channels, and achieve better delivery effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 is an architecture diagram of a multimedia material scheduling system provided by an embodiment of this specification;

[0015] Figure 2 is a flow chart of a multimedia material scheduling method provided by an embodiment of this specification;

[0016] Figure 3 is a schematic diagram of a data structure of a capacity unit provided by an embodiment of this specification;

[0017] Figure 4 It is a schematic diagram of a time period-flow rate change curve provided by an embodiment of this specification;

[0018] Figure 5is a flow chart of a note scheduling method for a social platform provided by an embodiment of this specification;

[0019] Figure 6 It is a flowchart of a processing process of applying a note scheduling method of a social platform in a capacity queue provided by an embodiment of the present specification;

[0020] Figure 7 It is a schematic diagram of a time period-flow corresponding change relationship curve provided by an embodiment of this specification;

[0021] Figure 8 It is a structural schematic diagram of a multimedia material scheduling device provided by an embodiment of this specification;

[0022] Fig. 9 It is a structural diagram of a note scheduling device for a social platform provided by an embodiment of this specification;

[0023] Fig.10 It is a structural block diagram of a computing device provided by one embodiment of this specification. DETAILED DESCRIPTION

[0024] Many specific details are described in the following description to facilitate a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of this specification, so this specification is not limited to the specific implementation disclosed below.

[0025] The terms used in one or more embodiments of this specification are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of this specification. The singular forms of "a", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0026] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0027] In addition, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0028] First, the terms involved in one or more embodiments of this specification are explained.

[0029] Multimedia materials are multimedia content used for dissemination and display, including but not limited to text, images, audio, video, etc.

[0030] At present, the scheduling strategy for delivering notes on social platforms is relatively simple, which is to set a total delivery amount for a period of time and deliver according to the constraints of the total delivery amount. However, this simple scheduling strategy often leads to congestion in delivery channels, and even the problem of delivery failure, resulting in poor delivery effect.

[0031] In order to avoid congestion in delivery channels, improve delivery completion and delivery effects, this specification provides a multimedia material scheduling method and a note scheduling method for a social platform. This specification also involves a multimedia material scheduling device, a note scheduling device for a social platform, a computing device, a computer-readable storage medium, and a computer program product, which are described in detail one by one in the following embodiments.

[0032] See also Figure 1 , Figure 1 The diagram shows an architecture diagram of a multimedia material scheduling system provided by an embodiment of the present specification. The multimedia material scheduling system may include: a server and a client.

[0033] The client is the client of the social platform and can be the sender of the multimedia material delivery request. The server can be the server of the social platform.

[0034] The client sends a multimedia material delivery request initiated by the user to the server.

[0035] The server, in response to receiving a multimedia material delivery request, determines a target project corresponding to the multimedia material delivery request and a delivery time range and a target delivery amount corresponding to the target project, determines a plurality of delivery time periods corresponding to the delivery time range according to a preset time period division rule, and determines delivery components of the target delivery amount dispersed to the plurality of delivery time periods, matches the delivery components corresponding to the plurality of delivery time periods with the remaining available delivery amounts of the corresponding capacity units, and obtains a matching result, wherein one capacity unit corresponds to one delivery time period, the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, the maximum available delivery amount is obtained based on the traffic predicted by the traffic prediction model for the corresponding delivery time period, and when it is determined that the target project is successfully allocated this time according to the matching results of the delivery components corresponding to the plurality of delivery time periods, the plurality of delivery components are allocated to the matching capacity units, and the multimedia materials of the delivery components are delivered to the promotion positions of the social platform within the delivery time period of the allocated capacity units.

[0036] After the multimedia materials of the target project are placed in the promotion position of the social platform, the browsing users of the social platform will browse the corresponding multimedia materials from the promotion position, thereby effectively increasing the exposure of the multimedia materials of the target project.

[0037] It should be noted that the client can be one or more clients. Multiple clients can establish communication connections through the server. The server can be used to provide multimedia material processing and transmission services between multiple clients. Multiple clients can act as senders or receivers respectively and communicate through the server.

[0038] For example, a user can interact with a server through a client to receive multimedia materials delivered by another client, or deliver multimedia materials to a promotional position on a social platform through a client, so that the multimedia materials are sent to another client through the server.

[0039] The client and the server are connected via a network. The network provides a medium for the communication link between the client and the server. The network can include various connection types, such as wired or wireless communication links or optical fiber cables. The data transmitted by the client may need to be encoded, transcoded, compressed, etc. before being published to the server.

[0040] The client can be a browser, an APP (Application), or a web application such as an H5 (HyperTextMarkup Language5, Hypertext Markup Language Version 5) application, or a light application (also known as a mini-program, a lightweight application) or a cloud application, etc. The client can be based on the software development kit (SDK, Software Development Kit) of the corresponding service provided by the server, such as based on the real-time communication (RTC, Real Time Communication) SDK development and acquisition. The client can be deployed in an electronic device, and needs to rely on the device to run or some APPs in the device to run. For example, the electronic device can have a display screen and support information browsing, such as a personal mobile terminal such as a mobile phone, a tablet computer, a personal computer, etc. Various other types of applications can also be configured in the electronic device, such as human-computer dialogue applications, model training applications, text processing applications, web browser applications, shopping applications, search applications, instant messaging tools, mailbox clients, social platform software, etc.

[0041] The server may include servers that provide various services, such as servers that provide communication services to multiple clients, servers that provide background training to support models used on clients, and servers that process data sent by clients. It should be noted that the server can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. The server may also be 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, content delivery networks (CDNs), and big data and artificial intelligence platforms, or an intelligent cloud computing server or intelligent cloud host with artificial intelligence technology.

[0042] It can be seen from the above system embodiments that the server in the system obtains the maximum available delivery amount of the capacity unit corresponding to the delivery period based on the traffic predicted by the traffic prediction model, so that the maximum available delivery amount of the capacity unit corresponds to the difference between high and low peak traffic in different time periods, and the target delivery amount of the multimedia materials of the target project is dispersed according to multiple delivery time periods, and the delivery components corresponding to the dispersed multiple delivery time periods are matched with the remaining available delivery amount of the capacity unit. When it is determined that the target project is successfully allocated according to the matching result, the delivery components are allocated to the matching capacity unit, thereby ensuring that the delivery is carried out under the condition of successful allocation, effectively ensuring the completion of the delivery, and making the dispersed delivery of multimedia materials accurately adapt to the difference between high and low peak traffic in different time periods, which can effectively avoid concentrated delivery, avoid congestion in delivery channels, and achieve better delivery effects.

[0043] See also Figure 2 , Figure 2 A flowchart of a multimedia material scheduling method provided by an embodiment of this specification is shown, which specifically includes the following steps:

[0044] Step 202: In response to receiving a multimedia material delivery request, determining a target item corresponding to the multimedia material delivery request and a delivery time range and a target delivery amount corresponding to the target item.

[0045] A multimedia material delivery request is a request for delivering multimedia materials. A multimedia material delivery request may carry the project information of the target project, the delivery time range and the target delivery volume. The target project may be understood as a delivery order created by the user to deliver multimedia materials, and the project information may be understood as delivery order information. The delivery time range is used to indicate the time range for delivering multimedia materials of the target project. The target delivery volume is the amount of traffic required for delivering the multimedia materials. Among them, the target delivery volume, i.e., the traffic volume, may be represented by any one or more indicators, such as: the number of visits, the number of unique visitors, the amount of data transmitted over the network, and so on.

[0046] For example, a user A creates an order 1 through the client to deliver 20 million traffic notes from 5:00 to 8:00 on a future day. The client then sends a request for the corresponding delivery note to the server of the social platform. The request carries the delivery time range "5:00 to 8:00 on a future day" and the target delivery volume "20 million" corresponding to the target item "Order 1".

[0047] Step 204: Determine a plurality of delivery time periods corresponding to the delivery time range according to a preset time period division rule, and determine delivery components of the target delivery amount distributed to the plurality of delivery time periods.

[0048] The preset time period division rule is a pre-set rule used to divide a time range into multiple time periods. For example, the preset time period division rule can be expressed as: division by 1 hour, division by 1 minute, division by 2 hours, etc. The delivery component is the delivery amount allocated to each time period after the target delivery amount is dispersed into multiple time periods. The sum of the delivery components corresponding to multiple time periods is the target delivery amount. The dispersion method can be set as needed, such as average dispersion, or dispersion according to a certain proportion, etc.

[0049] Exemplarily, assume that the preset time period division rule is: divided by 1 hour, the delivery time range is "5-8 o'clock on a certain day in the future", thus, the multiple time periods divided by "5-8 o'clock on a certain day in the future" include: 5-6 o'clock, 6-7 o'clock, and 7-8 o'clock.

[0050] Continuing with the above example, the target delivery amount of "20 million" is distributed to three time periods: 5-6 o'clock, 6-7 o'clock, and 7-8 o'clock. The corresponding delivery components can be 6 million: 7 million: 7 million.

[0051] In addition, in one or more embodiments, the target project also has a corresponding delivery quota. The delivery quota refers to the total traffic limit of the target project within the delivery time range. The delivery quota can be set by the user to which the target project belongs and confirmed to take effect by the social platform, or it can be set by the social platform and confirmed to take effect by the user, and this specification does not limit this. Therefore, before determining the delivery component of the target project, it is also possible to determine whether the target delivery amount exceeds the delivery quota of the target project. If it exceeds, there is no need to perform the step of determining the delivery component, and the target project creation failure information can be returned to the user. If it does not exceed, it can enter the step of determining multiple delivery time periods corresponding to the delivery time range according to the preset time period division rules. By setting the delivery quota for the target project, it is possible to more effectively manage and optimize the delivery of multimedia materials on the social platform, and ensure the fairness of resource allocation and user experience.

[0052] Step 206: Match the delivery components corresponding to the multiple delivery time periods respectively with the remaining available delivery quantities of the corresponding capacity units to obtain a matching result.

[0053] Among them, one capacity unit corresponds to one delivery period, and the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, and the maximum available delivery amount is obtained based on the traffic predicted by the traffic prediction model for the corresponding delivery period.

[0054] The traffic prediction model is a prediction model that predicts the traffic of the social platform corresponding to the input time period. The time period and the traffic prediction value can be in a one-to-one correspondence. The traffic can be represented by any one or more indicators, such as the number of visits, the number of unique visitors, the amount of network transmission data, etc.

[0055] A capacity unit corresponds to a time period. A capacity unit is a data structure used to manage the traffic of a social platform in a corresponding time period. The data structure of a capacity unit is not limited. For example, a capacity unit can be expressed as a capacity queue, a capacity linked list, etc. Information managed based on the capacity unit includes but is not limited to: the maximum available delivery of the capacity unit, the remaining available delivery information of the capacity unit and / or the allocated delivery information, etc. The remaining available capacity is equal to the difference between the maximum available delivery and the allocated delivery.

[0056] For example, Figure 3 FIG. 2 shows a schematic diagram of a data structure of a capacity unit provided by an embodiment of the present specification. Figure 3As shown, the capacity unit can be expressed as a data structure of capacity queue n, which can correspond to time period n and is used to manage the traffic of the social platform in time period n. The information managed based on capacity queue n includes: project information of projects with allocated delivery, such as "Order 1", "Order 2", "Order 3" and the traffic allocated to these projects, and the capacity upper limit of capacity queue n (that is, the maximum available delivery). Figure 3 As shown, the remaining available delivery volume is the part after deducting the delivery volume allocated to the queue from the maximum available delivery volume. For the newly created target item "Order 4", if the delivery component corresponding to "Order 4" in time period n is greater than the remaining available capacity, then the flow allocation of "Order 4" fails, which can also be understood as a failure to create the order. If the delivery component corresponding to "Order 4" in time period n is less than or equal to the remaining available capacity, then the flow allocation of "Order 4" can be successful, which can also be understood as a success in order creation.

[0057] The maximum available delivery volume of the capacity unit is obtained by the traffic prediction model based on the traffic predicted for the corresponding time period. Specifically, the traffic prediction model predicts future traffic based on the historical traffic data of the social platform. The specific method of obtaining the traffic prediction model is not limited, such as obtaining it based on statistics (such as time series analysis), based on machine learning (such as neural networks, support vector machines, etc.) training, or through other methods.

[0058] In one or more embodiments of the present specification, a time period-flow change curve is constructed based on historical flow data, and the flow corresponding to the input time period is predicted based on the time period-flow change curve. Specifically, the time period-flow change curve is obtained by the following steps:

[0059] Acquire historical traffic data of a social platform, the historical traffic data including: a plurality of historical time periods and true traffic values ​​of the social platform in the plurality of historical time periods;

[0060] Based on historical traffic data, only the time factor is considered to analyze the corresponding change relationship between time periods and traffic.

[0061] For example, the corresponding change relationship between time period and flow rate can be expressed by the following expression:

[0062]

[0063] Among them, y represents the flow rate, f(t) represents the function of the corresponding change relationship between the time period and the flow rate, a1 and a2 represent the slopes in the corresponding time period, t represents the time period, and b1, b2, and b3 represent constants.

[0064] Based on historical traffic data, only considering the time factor, the corresponding change relationship between time period and traffic is analyzed as follows: within a day, from 0:00 pm to t1:00 am, the traffic shows a fluctuating upward trend; from t1:00 am to t2:00 pm (such as 1 hour or 2 hours), it is at the peak of traffic, the traffic fluctuates, and generally remains at a high level; from t2:00 pm to 24:00 pm, the traffic shows a fluctuating downward trend. Based on this, we can construct the following Figure 4 The schematic diagram of the time period-flow rate change curve shows the change curve, where the horizontal axis is the time period and the vertical axis is the flow rate.

[0065] In one or more embodiments of this specification, the step of obtaining a traffic prediction model based on machine learning training may include:

[0066] Acquire historical traffic data of a social platform, the historical traffic data including: a plurality of historical time periods and true traffic values ​​of the social platform in the plurality of historical time periods;

[0067] The historical traffic data is used to train the traffic prediction model based on machine learning to obtain a trained traffic prediction model.

[0068] The above traffic prediction model can be a machine learning model, such as a large model, which can be trained with historical traffic data to perceive the changes in time series and real-time hot spots and other information, and give more accurate traffic prediction values. The traffic prediction model based on machine learning can be expressed by the following expression:

[0069] y=f(t,w1,w2,…,wn)

[0070] Where y represents the traffic prediction value, f(t,w1,w2,…,wn) represents the traffic prediction model, t represents the time period, and w1,w2,…,wn represent multiple features extracted by the model. The traffic prediction model makes predictions by extracting input features. For example, the extracted features include but are not limited to: time-related features, such as timestamps, days of the week, holidays, and any other factors that may affect traffic.

[0071] Step 208: When it is determined that the target project is successfully allocated this time according to the matching results of the delivery components corresponding to the multiple delivery time periods, the multiple delivery components are allocated to the matching capacity units.

[0072] The matching result is used to represent the result of matching the delivery components corresponding to multiple delivery time periods with the remaining available delivery quantity of the corresponding capacity unit. The matching result may include: successful matching or failed matching, or a numerical value indicating the degree of matching. A successful match means that the delivery component corresponding to the delivery time period is less than or equal to the remaining available delivery quantity of the corresponding capacity unit, and the delivery component and the capacity unit match; a failed match means that the delivery component corresponding to the delivery time period is greater than the remaining available capacity of the corresponding capacity unit, and the delivery component and the capacity unit do not match.

[0073] Based on this, the conditions for determining whether the current allocation of the target project is successful can be set according to the needs of the application scenario based on the matching results of the delivery components corresponding to multiple delivery time periods. For example: when the matching results of the delivery components of multiple delivery time periods all indicate a successful match, it can be determined that the current allocation of the target project is successful; when the matching results of the delivery components of more than a certain proportion of delivery time periods indicate a successful match, it can be determined that the current allocation of the target project is successful; when the matching results of the delivery components of any delivery time period indicate a failed match, it can be determined that the current allocation of the target project is a failed match; when the matching results of the delivery components of more than a certain proportion of delivery time periods indicate a failed match, it can be determined that the current allocation of the target project is a failed match.

[0074] If the target project allocation fails, the target project creation failure information can be returned to the target project user. It should be noted that the condition for determining whether the target project allocation is successful based on the matching result can be flexibly adjusted according to the scenario requirements, and this specification does not limit this.

[0075] Exemplarily, the target project A is dispersed into the delivery components A1, A2 and A3 (where T1 corresponds to A1, T2 corresponds to A2, and T3 corresponds to A3) of multiple delivery periods T1, T2 and T3, and matched with the remaining available delivery amounts of the capacity units U1, U2 and U3 (where T1 corresponds to U1, T2 corresponds to U2, and T3 corresponds to U3) corresponding to T1, T2 and T3 respectively, that is, A1 is matched with the remaining available delivery amount of U1, A2 is matched with the remaining available delivery amount of U2, and A3 is matched with the remaining available delivery amount of U3. When the delivery components A1, A2 and A3 are all less than or equal to the remaining available delivery amounts of the corresponding capacity units U1, U2 and U3, it is determined that the target project A is successfully allocated this time. If any of the delivery components A1, A2 and A3 does not match the remaining available capacity of the capacity unit to which it corresponds, the target project A fails to be allocated this time. Assuming that the allocation is successful, the delivery component A1 is allocated to the capacity unit U1, the delivery component A2 is allocated to the capacity unit U2, and the delivery component A3 is allocated to the capacity unit U3.

[0076] Step 210: placing the multimedia material of the delivery component into the promotion position of the social platform within the delivery period of the allocated capacity unit.

[0077] Social platforms can be understood as online services or applications that provide users with communication, sharing and interaction through network connections. Through social platforms, users can communicate and establish connections with other users through text, pictures, videos, etc.

[0078] The content on the social platform is in various forms, including short videos, pictures and texts, etc. In one or more embodiments, the multimedia material can be a note, that is, the short videos, pictures and texts on the social platform are presented through notes, and users can record and share content through notes to form interactions.

[0079] Promotional position refers to the position for placing multimedia materials determined based on the promotion strategy of the social platform. In the social platform, the promotion strategy used to determine the promotion position of multimedia materials is not limited, and it is set according to the specific needs of the application scenario. For example: In one promotion strategy, recommendations and promotions are made based on the browsing interests of users. Specifically, the multimedia materials can be matched with the browsing information of platform users to find out the user group that is interested in the multimedia materials to be placed, and the multimedia materials can be accurately placed on the browsing page of the user group. At this time, the browsing page of the interested user group is the advertising position. Another example: In another promotion strategy, recommendations and promotions are made based on the search behavior of users. Specifically, when users search for keywords, the multimedia materials can be matched with the keywords searched by the users to find out the search users who search for keywords related to the multimedia materials, and the multimedia materials can be accurately placed on the search results interface of the search users. At this time, the search results page is the advertising position.

[0080] Using the previous example, the delivery component A1 is assigned to the capacity unit U1, the delivery component A2 is assigned to the capacity unit U2, and the delivery component A3 is assigned to the capacity unit U3. Therefore, the notes of the delivery component A1 of the target project A are delivered during the T1 period; the notes of the delivery component A2 of the target project A are delivered during the T2 period; and the notes of the delivery component A3 of the target project A are delivered during the T3 period.

[0081] It can be seen that since this method obtains the maximum available delivery volume of the capacity units corresponding to multiple delivery time periods based on the traffic predicted by the traffic prediction model, the maximum available delivery volume of the capacity unit corresponds to the peak and low peak differences in different time periods, and the target delivery volume of the multimedia materials of the target project is dispersed according to multiple delivery time periods, and the delivery components corresponding to the dispersed multiple delivery time periods are matched with the remaining available delivery volume of the capacity unit. When it is determined that the target project is successfully allocated this time according to the matching result, it is allocated to the matching capacity unit, thereby ensuring that the delivery is carried out under successful allocation, effectively ensuring the completion of the delivery, and making the dispersed delivery of multimedia materials accurately adapt to the peak and low peak differences in different time periods, which can effectively avoid concentrated delivery, avoid congestion in delivery channels, and achieve better delivery effects.

[0082] In the embodiments of this specification, the capacity unit is a data structure used to manage the traffic of the social platform in the corresponding time period. A traffic statistics cycle will be divided into multiple time periods, and a capacity unit corresponds to a time period. Therefore, the capacity units created for each time period of a traffic statistics cycle can still be used in the next traffic statistics cycle as long as the traffic statistics cycle and time period remain unchanged, and repeat over and over again. In actual applications, according to the needs of the scenario, or triggered by a preset event, a capacity unit can be recreated for a traffic statistics cycle, and the maximum available delivery volume corresponding to the capacity unit can be re-predicted. For example, before a new traffic statistics cycle, the traffic of the upcoming new traffic statistics cycle can be predicted, and capacity units can be created in time periods. For another example, when the resources of the social platform are adjusted, resulting in a change in the traffic quota, the traffic forecast can be re-performed and the capacity units can be created in time periods.

[0083] Therefore, in one or more embodiments of this specification, before step 206, the method further includes the steps of creating a capacity unit and predicting the maximum available delivery volume of the capacity unit. Specifically, before step 206, the method further includes:

[0084] Determine target time frame;

[0085] According to the preset time period division rule, determining a plurality of time periods to be tested corresponding to the target time range;

[0086] Creating corresponding capacity units for the multiple time periods to be tested, respectively, wherein the capacity unit corresponding to the time period to be tested is consistent with the capacity unit corresponding to the delivery time period belonging to the same time period;

[0087] Using the traffic prediction model, predict the traffic corresponding to any time period to be tested on the social platform to obtain a traffic prediction value corresponding to any time period to be tested;

[0088] According to the traffic forecast value corresponding to any time period to be tested, the maximum available delivery volume of the capacity unit corresponding to any time period to be tested is determined.

[0089] The target time range may have a corresponding traffic quota. The traffic quota refers to the upper limit of the total traffic of the social platform within the target time range. The target time range is the range between the start time and the end time of a traffic statistics cycle. For example, if a traffic statistics cycle is one month, the target time range may be the range between the first day and the last day of the month. The time period to be tested can be understood as the time period when the social platform receives visits.

[0090] Assume that the target time range is expressed as T(1)–T(N) (such as any time range such as a single month or a double month), and the traffic quota is expressed as Quota_Total. Divide the target time range into N test time periods according to the preset time period division rules (such as any time period division rules such as days, hours, minutes, etc.), then the i-th time period can be expressed as T(i)–T(i+1), where T(i) is the start time of the i-th time period and T(i+1) is the end time of the i-th time period.

[0091] Exemplarily, the target time range is one month, and the preset time period division rule is to divide by 1 hour. Then the multiple time periods to be tested corresponding to the target time range include: every hour from the first day to the last day of a month, and one hour is a time period to be tested.

[0092] The capacity unit corresponding to the time period to be tested is consistent with the capacity unit corresponding to the delivery time period belonging to the same time period. Continuing with the previous example, assuming that the multiple delivery time periods of the target project are 5-6 o'clock, 6-7 o'clock, and 7-8 o'clock, the three time periods to be tested 5-6 o'clock, 6-7 o'clock, and 7-8 o'clock correspond to capacity unit n, capacity unit n+1, and capacity unit n+2, respectively. It can be seen that these three delivery time periods and the three time periods to be tested actually belong to the same time period, and all correspond to capacity unit n, capacity unit n+1, and capacity unit n+2.

[0093] When creating corresponding capacity units for multiple time periods to be tested, the initial remaining available capacity of the capacity unit can be consistent with the maximum available delivery amount. One capacity unit corresponds to one time period, based on the information managed by the capacity unit, including but not limited to: the maximum available delivery amount of the capacity unit, the remaining available delivery amount information of the capacity unit and / or the allocated delivery amount information, etc. Exemplarily, the data structure of the created capacity unit can be as follows: Figure 3 The data structure of the capacity queue is shown.

[0094] The structure of the capacity queue and the implementation of the traffic prediction model can refer to the description of the aforementioned embodiments and will not be repeated here. In one or more embodiments, prediction is performed with the traffic quota as a constraint condition, so that the sum of the traffic prediction values ​​corresponding to the multiple time periods to be tested is less than or equal to the traffic quota. Specifically, when using the traffic prediction model for prediction, the traffic quota can be used as the input of the model and input into the model so that the model can make predictions with the traffic quota as a constraint condition.

[0095] In the above embodiment, traffic units are created in different time periods, and a traffic prediction model is used to predict the traffic prediction value corresponding to the time period as the maximum available delivery amount of the traffic unit, thereby ensuring that the traffic managed by the traffic unit does not exceed the upper limit of the platform and corresponds to the traffic difference between the peak and off-peak periods of the platform, so that accurate traffic allocation can be performed subsequently according to the remaining available delivery amount of the traffic unit.

[0096] In order to more accurately predict the maximum available delivery volume corresponding to the capacity unit, in one or more embodiments of this specification, the traffic prediction model is used to predict the traffic corresponding to the social platform in any time period to be tested, and the traffic prediction value corresponding to any time period to be tested is obtained, including:

[0097] Input any time period to be measured and flow influencing factor data into the flow prediction model, and use the flow prediction model to predict the corresponding flow prediction value of any time period to be measured under the influence of the flow influencing factor data;

[0098] Among them, the traffic prediction model is obtained based on the historical traffic data of the social platform, and the historical traffic data includes: multiple historical time periods belonging to the same time period as any time period to be measured, historical traffic influencing factor data corresponding to at least one of the historical time periods, and the true value of the traffic of the social platform in the multiple historical time periods.

[0099] In one or more embodiments, multiple time periods to be measured and flow influencing factor data can be input into the flow prediction model at one time, and the flow prediction model predicts the flow prediction values ​​corresponding to the multiple time periods to be measured under the influence of the flow influencing factor data. In addition, the flow quota can also be input into the flow prediction model so that the sum of the flow prediction values ​​corresponding to the multiple time periods to be measured is less than or equal to the flow quota.

[0100] For example, assuming that a live broadcast opening event will be held at a certain moment in the future, the traffic of the social platform within the live broadcast period will increase significantly. In order to allow more traffic to be used for delivery within the live broadcast period, the traffic influencing factor data of "live broadcast event" can be input into the traffic prediction model to obtain the maximum available delivery within the live broadcast period under the influence of "live broadcast event", and the size of the maximum available delivery of the capacity unit is updated accordingly, so that the social platform can better undertake this part of the increased traffic. Similarly, assuming that an "energy saving and emission reduction, put down your phone for 1 hour" event will be held at a certain moment in the future. Under this event, the traffic of the social platform will decrease to a certain extent. In order to avoid the event causing the delivery to fail to be completed, the traffic influencing factor data of "energy saving and emission reduction, put down your phone for 1 hour" can be input into the traffic prediction model to obtain the maximum available delivery within the activity period under the influence of "energy saving and emission reduction, put down your phone for 1 hour", and the size of the maximum available delivery of the capacity unit is updated accordingly, reducing the remaining available delivery, reducing the delivery within the activity period, and avoiding the problem of delivery failing to be completed.

[0101] In the historical traffic data of the above embodiment, the true traffic value corresponding to at least one historical period is the true traffic value corresponding to the social platform under the influence of the corresponding historical traffic influencing factor data. Traffic influencing factor data is data related to factors that affect the number of visits to the social platform. Based on the input of historical traffic influencing factor data, the model can learn the pattern of the impact of historical traffic influencing factor data on traffic, and more accurately predict the traffic prediction value of a certain period.

[0102] In one or more embodiments, the traffic prediction model based on machine learning can be used to predict the traffic flow of the target time period to be tested. Accordingly, the input during training includes: historical traffic data of the target time period to be tested. For example, the historical traffic data of the target time period to be tested can include: the true value of the traffic flow of the same target time period to be tested in the last traffic statistical cycle, the true value of the traffic flow of the same target time period to be tested last week, the true value of the traffic flow of the same target time period to be tested last month, the true value of the traffic flow of the same target time period to be tested last quarter, and the true value of the traffic flow of the same target time period to be tested last year. Furthermore, the traffic impact factor data can be expressed as the impact value corresponding to the external event impact factor, such as the impact value can generally be taken as a conventional value of 1, and can be adjusted to a larger or smaller value according to the received adjustment instruction as needed. When adjusted to a larger value, such as 1.01, 1.02, etc., it can be indicated that the event will cause an increase in platform traffic; when adjusted to a smaller value, such as 0.99, 0.98, etc., it can be indicated that the event will cause a decrease in platform traffic. In addition, in order to make the prediction results of the traffic prediction model more accurate, the interaction data of newly released notes in the adjacent time period of the target time period, the interaction data of old released notes, and / or the interaction data of the notes of the top bloggers can also be input into the traffic prediction model, so that the traffic prediction model can refer to the interaction data to more accurately infer the traffic of the target time period. For example, the adjacent time period can include: in the x adjacent time periods before the target time period, the interaction data includes the number of new notes released, the number of likes, the number of collections, the number of feedbacks, etc., and / or the number of old notes released, the number of likes, the number of collections, etc. The top blogger can be understood as a blogger whose number of fans is greater than a preset fan number threshold.

[0103] The reasoning process of the model can be understood as: taking the true value of the flow in the same historical period as the learning object, combining the influence of the flow influencing factor data, fitting the flow prediction value of the target test period to be predicted and outputting it.

[0104] After creating corresponding capacity units for the multiple time periods to be tested, the management information of the capacity units can also be recorded, and in the subsequent process, the management information can be updated according to the allocation situation, so as to perform accurate flow allocation according to the management information of the flow unit. Specifically, after creating corresponding capacity units for the multiple time periods to be tested, the method further includes:

[0105] In the management information of the capacity unit, the information of the time period to be measured, the maximum available delivery amount information and the flow usage information corresponding to the capacity unit are recorded, and the flow usage information includes the remaining available delivery amount information and / or the allocated delivery amount information;

[0106] The allocating the multiple delivery components to matching capacity units includes:

[0107] In the management information of the capacity unit, the project information of the allocated target project, the allocated delivery component size, and the delivery status information of the multimedia material of the delivery component are recorded, and the traffic usage information is updated according to the size of the delivery component.

[0108] The delivery status information is information used to indicate the delivery status of the multimedia material of the allocated delivery component. For example, the delivery status information may include: pending delivery, delivery in progress, or completed. Furthermore, the delivery status information may also include more detailed information, such as: delivery time details (e.g., assuming that the time period is divided into hours, the delivery time details can be further accurate to minutes), delivery target user information (e.g., it can be the target audience label of the multimedia material), delivery performance information (e.g., it can be click-through rate, conversion rate, etc.).

[0109] For example, assuming that capacity unit n is allocated with the delivery components of order 1, order 2, and order 3, based on this, the management information recorded for capacity unit n corresponding to time period n may include the content contained in the following table:

[0110]

[0111] The above embodiment achieves more refined traffic allocation and multimedia material delivery management by recording the maximum available delivery amount information in the management information of the capacity unit, updating the traffic usage information according to the size of the allocated delivery component, and recording the delivery status information of the multimedia material, thereby achieving accurate delivery and improving the delivery effect. For example: the usage level of the capacity unit can be monitored according to the traffic usage information recorded in the management information of the capacity unit. If it is monitored that the usage level of any one or more capacity units is low, a notification can be sent to the delivery user to prompt the delivery user to use the time period of these capacity units for delivery to avoid wasting traffic.

[0112] In order to further improve the delivery completion rate, in one or more embodiments of this specification, determining the delivery components of the target delivery amount distributed to the multiple delivery time periods includes:

[0113] The target delivery amount is divided according to the ratio of the maximum available delivery amounts of the capacity units respectively corresponding to the multiple delivery time periods, so as to determine the delivery components respectively corresponding to the multiple delivery time periods.

[0114] Exemplarily, assuming that the maximum available delivery amounts of the capacity units corresponding to multiple delivery time periods are 12 million, 14 million, and 14 million, respectively, and the ratio of the three is 1.2:1.4:1.4, based on this, the target delivery amount of 20 million is divided according to this ratio, and the resulting delivery components are: 6 million, 7 million, and 7 million.

[0115] By dividing the total target delivery volume according to the ratio of the maximum available delivery volume of the corresponding capacity unit, the matching degree between the divided delivery components and the remaining available delivery volume of the capacity unit can be improved, thereby effectively improving the probability of successful matching and the delivery completion degree.

[0116] After determining the delivery components corresponding to multiple delivery segments, when matching the delivery components with the remaining available delivery quantities of the corresponding capacity units, it is necessary to at least match the size of the delivery components with the size of the remaining available delivery quantities of the capacity units to avoid the amount of multimedia materials delivered exceeding the access volume of the corresponding time period.

[0117] Furthermore, in some embodiments, other additional matching conditions may be combined for matching to make the matching more stringent or more relaxed according to the application scenario. Specifically, the delivery components corresponding to the multiple delivery time periods are matched with the remaining available delivery amounts of the corresponding capacity units, including:

[0118] Get additional matching conditions for the target item;

[0119] According to the additional matching conditions, the delivery components corresponding to the multiple delivery time periods are matched with the remaining available delivery quantities of the corresponding capacity units.

[0120] Additional matching conditions refer to looser conditions or stricter conditions that are flexibly set for matching when matching the delivery component with the remaining available capacity of the corresponding capacity unit. Among them, looser conditions are used to improve the matching success rate to avoid matching failures caused by the delivery component being larger than the remaining available capacity of the corresponding capacity unit; stricter conditions are used for accurate matching to ensure the delivery effect. In actual applications, additional matching conditions can be set for the target project according to the needs of the scenario, or set for the target project according to user needs, and this manual does not limit this. For example: additional matching conditions can be saved on the server as user configuration information, or carried in the delivery request when the user initiates the delivery request. Therefore, the additional matching conditions for the target project can be obtained from the configuration information saved by the server, or from the user's delivery request.

[0121] In one or more embodiments of this specification, the additional matching condition may include: any one or more of the following conditions:

[0122] Matching conditions for matching capacity units corresponding to a specified target period;

[0123] A matching condition for matching according to the priority information of the target project;

[0124] The start time and the end time of the target delivery period are adjusted, and the duration of the target delivery period before and after the adjustment remains unchanged. The target delivery period is any one or more delivery periods among the multiple delivery periods.

[0125] For example, the matching condition for matching the capacity unit corresponding to the specified target period can be expressed as: allowing the capacity unit corresponding to a certain extended period to be matched (more relaxed); or requiring the capacity unit corresponding to a certain peak period to be matched (more strict). Since the flow size or time period of the capacity unit corresponding to the specified target period is more in line with the delivery requirements, this implementation method can effectively improve the user experience or increase the matching success rate.

[0126] For another example, the matching condition for matching according to the priority information of the target project may be expressed as: a project with a higher priority has priority allocation. Since matching is performed according to the priority information of the target project, the matching success rate of the target project with a higher priority can be effectively guaranteed.

[0127] For another example, the delivery period is adjusted according to the time period adjustment strategy, and the matching condition is matched according to the adjusted delivery period, which can be expressed as: adjusting the delivery start and end time of the target delivery period within the range of x hours, and ensuring that the total delivery duration of the target delivery period is still the original delivery duration. This implementation method is suitable for scenarios where the target project is not sensitive to business peaks and low peaks, and the multimedia materials of the target project are allowed to be delivered at any time period.

[0128] Through the above embodiments, the usage of the capacity unit is matched in combination with the additional matching conditions of the target project, so as to perform a looser or stricter match for the target project, which can effectively improve the delivery completion degree or the delivery accuracy.

[0129] In order to allow the user to timely understand the congestion and / or completion status of the multimedia materials to be delivered, in one or more embodiments of this specification, the method further includes:

[0130] Obtaining the allocated delivery amount of the capacity unit to which the delivery component is to be allocated;

[0131] According to the delivery component and the allocated delivery amount, congestion prediction and / or delivery completion prediction are performed on the capacity unit to obtain a congestion prediction result and / or delivery completion prediction result of the capacity unit;

[0132] The congestion prediction result and / or delivery completion prediction result are sent to the user of the target project.

[0133] The specific implementation methods of the predicted congestion data and / or predicted delivery completion data are not limited. For example, they may include:

[0134] Determine the delivery components corresponding to the multiple delivery periods of the target project, and the capacity units corresponding to the delivery components;

[0135] Obtain the allocated delivery amount of the capacity unit corresponding to each delivery component;

[0136] For each capacity unit, the allocated delivery amount of the capacity unit and the delivery component to be allocated to the capacity unit by the target project are added together to obtain the traffic usage forecast value of the capacity unit;

[0137] If the traffic usage forecast value of any capacity unit exceeds the maximum available delivery volume of any capacity unit, the congestion forecast result of any capacity unit is congestion, otherwise the congestion forecast result is not congestion;

[0138] Add up the excess traffic of all delivery components of the target project that exceeds the maximum available delivery volume of the corresponding capacity unit to obtain the total traffic that cannot be delivered;

[0139] Based on the sum of all delivery components of the target project and the sum of the traffic that cannot be delivered, the delivery completion rate is obtained as the delivery completion prediction result.

[0140] For example, assuming that there is a target project A, the delivery components of the target project A1, A2 and A3, the corresponding capacity units U1, U2 and U3, the allocated delivery amounts of the capacity units U1, U2 and U3 are P1, P2 and P3, and the maximum available delivery amounts are S1, S2 and S3, then the congestion prediction result and the delivery completion prediction result can be determined by the following calculation:

[0141] If A1+P1>S1, the congestion prediction result of A1 is congestion, otherwise it is not congested;

[0142] If A2+P2>S2, the congestion prediction result of A2 is congestion, otherwise it is not congested;

[0143] If A3+P3>S3, the congestion prediction result of A3 is congestion, otherwise it is not congested;

[0144] Assuming A1+P1>S1, the delivery completion rate = (A1+P1-S1) / (A1+A2+A3).

[0145] It should be noted that the above-mentioned steps of predicting congestion data and / or predicting delivery completion data can be performed before allocating multiple delivery components to matching capacity units in step 208, or can be performed after step 208. In the case of being performed before step 208, the user can timely understand the congestion and / or completion status of the multimedia material before determining to deliver the multimedia material to the matching capacity unit, so as to choose to continue delivery, or re-initiate the request after updating the demand. In the case of being performed after step 208, the user can timely track the congestion and / or completion status, thereby improving the user experience.

[0146] According to the above embodiment, in one or more embodiments of this specification, before allocating the multiple delivery components to the matching capacity units, the method further includes:

[0147] Receiving delivery confirmation information returned by the user in response to the congestion prediction result and / or delivery completion prediction result;

[0148] When it is determined according to the delivery confirmation information that the user confirms to execute the delivery, the step of allocating the multiple delivery components to matching capacity units is entered.

[0149] In the above embodiment, for the part of the delivery that exceeds the capacity unit, after being prompted to the user, the user can choose to continue the delivery, indicating that he or she is willing to bear the risk of not being able to complete the delivery 100%, or the user can choose to refuse to continue the delivery. The reason for allowing the user to choose to continue the delivery is that the maximum available delivery amount is the possible difference between the model's prediction result and the actual delivery, and there is a probability that the actual delivery can be completed. For example: in actual applications, the maximum available delivery amount of each capacity unit can be predicted before the delivery of the capacity unit begins. The size of the maximum available delivery amount of the capacity unit will increase or decrease or remain unchanged due to the prediction result of the traffic prediction model. After the capacity unit starts to be delivered, the maximum available delivery amount of the capacity unit is the traffic prediction value output by the model and remains unchanged until the next model prediction is made.

[0150] Corresponding to the above method embodiment, this specification also provides a note scheduling method for a social platform. Figure 5 , Figure 5 A flowchart of a note scheduling method for a social platform provided by an embodiment of the present specification is shown, which specifically includes the following steps:

[0151] Step 502: In response to receiving a note delivery request, determining a target project corresponding to the note delivery request and a delivery time range and a target delivery amount corresponding to the target project.

[0152] Notes are multimedia materials, including but not limited to text, images, audio, video, etc. Note delivery request is a request to deliver notes, which can carry the project information of the target project, delivery time range and target delivery volume.

[0153] Step 504: Determine a plurality of delivery time periods corresponding to the delivery time range according to a preset time period division rule, and determine delivery components of the target delivery amount distributed to the plurality of delivery time periods.

[0154] Step 506: Match the delivery components corresponding to the multiple delivery time periods with the remaining available delivery quantities of the corresponding capacity units to obtain a matching result.

[0155] Among them, one capacity unit corresponds to one delivery period, and the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, and the maximum available delivery amount is obtained based on the traffic predicted by the traffic prediction model for the corresponding delivery period.

[0156] Step 508: When it is determined that the target project is successfully allocated this time according to the matching results of the delivery components corresponding to the multiple delivery time periods, the multiple delivery components are allocated to the matching capacity units.

[0157] Step 510: The notes of the delivery component are delivered to the promotion position of the social platform within the delivery period of the allocated capacity unit.

[0158] It should be noted that the specific implementation details of the above steps 502-510 can refer to the description of one or more of the above embodiments, such as the description of steps 202-210, and this embodiment will not be repeated in detail here.

[0159] Since this method obtains the maximum available delivery volume of the capacity units corresponding to multiple delivery time periods based on the traffic predicted by the traffic prediction model, the maximum available delivery volume of the capacity unit corresponds to the peak-peak difference of different time periods, and the target delivery volume of the notes to be delivered to the target project is dispersed according to multiple delivery time periods, and the delivery components corresponding to the dispersed multiple delivery time periods are matched with the remaining available delivery volume of the capacity unit. When it is determined that the target project is successfully allocated this time according to the matching result, the delivery components are allocated to the matching capacity units, thereby ensuring that the delivery is carried out under the condition of successful allocation, effectively ensuring the completion of the delivery, and making the dispersed delivery of the notes accurately adapt to the peak-peak difference of different time periods, which can effectively avoid concentrated delivery, avoid congestion in delivery channels, and achieve better delivery effect.

[0160] See also Figure 6 , Figure 6A flowchart showing a processing process of applying a note scheduling method of a social platform provided by an embodiment of the present specification in a capacity queue is shown, which specifically includes the following steps:

[0161] Step 602: Determine a target time range and a plurality of forecast time periods within the target time range.

[0162] In one or more embodiments, the target time range T(1)-T(N) may be determined based on the traffic statistics cycle, where one traffic statistics cycle is a target time range. For example, the target time range may include a traffic statistics cycle of any length, such as a single month or two months.

[0163] The multiple time periods to be tested can be obtained by dividing the target time range according to the preset time period division rule. The preset time period division rule may include: dividing by granularity such as day, hour, minute, etc. Assuming that the target time range is divided into N time periods to be tested, the i-th time period to be tested is T(i)–T(i+1).

[0164] Step 604: Establish a traffic prediction model.

[0165] The traffic prediction model can establish a traffic fluctuation model based on the platform's traffic fluctuations over time periods.

[0166] For example, when the time period to be tested is divided into 1 hour, within a day, generally from 0:00 p.m. to t1:00 a.m., the flow rate shows a fluctuating upward trend. From t1:00 a.m. to t2:00 p.m., the flow rate is at a peak period, and the flow rate fluctuates but generally remains at a high level. From t2:00 p.m. to 24:00 a.m., the flow rate shows a fluctuating downward trend. Based on the corresponding change relationship between the time period and the flow rate, a flow prediction model as provided in the above embodiment can be constructed, that is, a function f(t) representing the corresponding change relationship between the time period and the flow rate.

[0167] For another example, based on machine learning models, such as large models, we can perceive the patterns of changes in time periods and traffic flow, and combine real-time hot spots and other information to predict more accurate traffic flow prediction values ​​for the time period to be tested.

[0168] Exemplarily, in an application scenario, the effect of delivery is evaluated with a single month as the traffic statistics cycle, so a single month is used as the target time range. Assume that the monthly exposure of the social platform is 10 billion, and 10% of the exposure of the market is used to deliver notes. Therefore, the total amount of delivery for the delivery of notes is 1 billion, and the daily delivery amount (assuming an average distribution) is 1 billion / 30 days = 33.3 million / day. Assume that the social platform has three main delivery users A, B, and C. The total delivery amount of a traffic statistics cycle is distributed to the three delivery users, and each delivery user is allocated a delivery amount of 200 million, 500 million, and 300 million.

[0169] For the above application scenario, in order to predict the maximum available delivery volume based on the capacity queue, a traffic prediction model must be built first. Since the target time range is a single month, the preset time division rule is divided by hour, and the total delivery amount within the target time range is 1 billion, a traffic prediction model can be built based on historical traffic data when only the time factor is considered. For example, the following can be obtained based on historical traffic data: Figure 7 The time period-flow corresponding change relationship curve shown in the figure shows that the flow fluctuates with the time period, with high and low peaks. Figure 7 The time period-flow corresponding change relationship curve shown in the figure can construct the flow-time period fluctuation model (that is, the flow prediction model) shown in the following expression:

[0170]

[0171] Here, t represents the time period.

[0172] Based on the above traffic-time fluctuation model, the maximum available delivery volume for each hour in a single month can be determined.

[0173] Step 606: Create a capacity queue corresponding to the time period to be measured.

[0174] For example, a corresponding capacity queue can be established for each time period to be tested. The management information corresponding to a capacity queue may include: the corresponding time period to be tested, the maximum available delivery volume of the queue, the allocated delivery volume of the queue, etc.

[0175] After the capacity queue is created through the above steps 602 to 606, the created capacity queue can be registered to the capacity manager. The capacity manager manages the registered capacity queue. For example, the capacity manager can receive the scheduling request of the capacity scheduler, manage all capacity queues based on the scheduling request, and record the management information of the capacity queue, such as the created order and delivery status information.

[0176] Exemplarily, following the previous example, a capacity queue can be created for each hour in a single month, and the maximum available delivery of the capacity queue is obtained according to the traffic predicted by the traffic prediction model for the corresponding prediction period. According to the above traffic-time period fluctuation model, the maximum available delivery of each capacity queue is predicted. For example, taking 24 hours on a certain day in a single month as an example, the maximum available delivery of the capacity queues corresponding to the 24 prediction periods are: 20 million (0-1 o'clock), 40 million (1-2 o'clock), 60 million (2-3 o'clock), 80 million (3-4 o'clock), 10 million (4-5 o'clock), 12 million (5-6 o'clock), 14 million (6-20 o'clock), 112.5 million (20-21 o'clock), 85 million (21-22 o'clock), 57.5 million (22-23 o'clock), and 30 million (23-24 o'clock).

[0177] It should be noted that the maximum available delivery volume of the capacity queue can be determined based on the predicted value of the traffic prediction model at the initial moment, and then, according to the needs of the actual application scenario, there can be a certain range of change based on the predicted value to more flexibly respond to changes in actual conditions.

[0178] After the capacity queue is created, the following steps 608-620 can be performed to schedule note delivery for users based on the capacity queue.

[0179] Step 608: The client generates an order creation request.

[0180] For example, the client can generate an order creation request in response to a user's order creation operation, or can automatically generate an order creation request in response to other triggering events. An order creation request is a request used to trigger the creation of a delivery order. A delivery order is an order used to deliver a note to a social platform.

[0181] Notes are a form of multimedia material on some social platforms. Notes can contain rich text, pictures, and videos. In some activities on social platforms, such as preheating the volume, supporting bloggers with traffic, and conducting targeted cooperation with products or merchants, users or platforms may have the need to create and place orders for notes in order to promote the notes on social platforms and make the notes reach the expected popularity.

[0182] Step 610: The client sends an order creation request to the capacity scheduler of the server.

[0183] Step 612: In response to receiving the order creation request, the capacity scheduler determines the delivery time range and target delivery quantity corresponding to the delivery order to be created.

[0184] In social platforms, there can be multiple users who can initiate note delivery requests. Social platforms can allocate corresponding delivery quotas to users according to the needs of the scenario. For example: within the target time range, according to the needs of the scenario, the total delivery quota Quota_Total is allocated to each user, and each user is allocated the corresponding delivery quota. When a user initiates an order creation request, the target delivery volume corresponding to the delivery time range should be within the delivery quota.

[0185] Exemplarily, at any time t0 within the target time range, multiple users have respectively initiated order creation requests. The order creation request carries the corresponding order information, such as: the delivery start time t1 (t1>=t0), the end time t2 (t2>t1) and the target delivery amount quota, and there may also be other supplementary information. If a user's delivery quota within the current target time range has been consumed, that is, the notes with the delivery quota have been delivered, then within the current target time range, the user can no longer initiate an order creation request. If the notes that have been delivered within the current target time range have not consumed the delivery quota, you can continue to initiate order creation requests.

[0186] The capacity scheduler receives the request to create a delivery order, divides the target delivery volume corresponding to the delivery order in the following step 608, obtains the delivery components of the delivery order, and then matches the delivery components with the remaining available delivery volume of the capacity queue in the following step 610. If the target item is successfully allocated this time according to the matching result, the order is successfully created; if the matching fails, the order fails to be created.

[0187] Step 614: The capacity scheduler determines a plurality of delivery time periods corresponding to the delivery time range according to a preset time period division rule, and determines delivery components of the target delivery amount distributed to the plurality of delivery time periods.

[0188] For example, if the time period to be tested is divided into 1 hour, the capacity scheduler can also divide the delivery time range of the delivery order into 1 hour and determine multiple delivery time periods. The target delivery volume is dispersed according to multiple delivery time periods, and the delivery components corresponding to each delivery time period are determined.

[0189] When dispersing the target delivery volume according to multiple delivery time periods, it can be divided evenly or in different proportions. For example: the total target delivery volume is quota, and the delivery order spans a total of n delivery time periods, then the delivery component corresponding to each delivery time period is quota / n. Another example: based on the traffic difference between peak and off-peak time periods, if the delivery order spans peak and off-peak time periods, the target delivery volume can be divided according to a certain proportion, so that the delivery components dispersed during peak and off-peak time periods are different. Furthermore, the delivery volume can be divided based on the proportion of the maximum available delivery volume of the capacity queue corresponding to the time period. In this way, the delivery component is large during peak time periods and small during off-peak time periods. The proportion of the divided delivery components corresponds to the maximum available delivery volume of the capacity queue, and the delivery effect is better.

[0190] After dividing the target delivery volume of the delivery order, a delivery component list of the delivery order can be obtained. The delivery component list lists the corresponding relationship between the delivery time period and the delivery component.

[0191] For example, suppose that user A creates delivery order 1, and the order content of delivery order 1 is to deliver 20 million notes between 5 and 8 o'clock on a certain day in the future, and the total delivery quota of user A is 200 million. The capacity scheduler determines that the remaining usage quota of user A is 200 million, which is greater than 20 million, and determines to allow the creation of delivery order 1. Then, the 20 million is distributed to the three time periods of 5-6, 6-7, and 7-8 in proportion, with a ratio of 1.2:1.4:1.4, and the resulting delivery components are 6 million, 7 million, and 7 million.

[0192] Step 616: The capacity scheduler matches the delivery components corresponding to the multiple delivery time periods with the remaining available delivery quantities of the corresponding capacity queues to obtain a matching result.

[0193] The capacity scheduler may obtain the remaining available capacity of the capacity queues corresponding to the multiple delivery time periods from the capacity manager, and match the delivery components in the delivery component list with the obtained remaining available capacity of the capacity queues of the corresponding time periods.

[0194] Step 618: When the capacity scheduler determines that the delivery order is successfully allocated this time based on the matching results of the delivery components corresponding to the multiple delivery time periods, the capacity scheduler allocates the multiple delivery components to the matching capacity queues.

[0195] For example, assume that there is a capacity queue n. Before order 4 is created, the delivery components of order 1, order 2, and order 3 have been allocated to capacity queue n. If after the delivery components of order 4 are allocated to capacity queue n, the total amount of the allocated delivery of capacity queue n will exceed the maximum available delivery of capacity queue n, then the matching result of the delivery component is a match failure. If after the delivery components of order 4 are allocated to capacity queue n, the total amount of the allocated delivery of capacity queue n is still less than or equal to the maximum available delivery of capacity queue n, then the matching result of the delivery component is a match success.

[0196] In addition, when it is determined that the allocation of the delivery order has failed this time, re-matching can be performed in combination with additional matching conditions (such as more relaxed additional matching conditions provided by the business party).

[0197] Among them, the additional matching conditions can be determined based on the supplementary information added by the user when creating the delivery order. For example, the additional matching conditions can be expressed as: allowing matching in a capacity queue for a certain extended period; requiring matching during business peak hours; matching according to the priority of the delivery order, etc. The purpose of setting additional matching conditions is to make the matching process between the delivery component of the delivery order and the capacity queue more stringent or more relaxed as needed to meet the needs of the scenario.

[0198] The capacity scheduler can match the remaining available delivery volume of the capacity queue in the capacity manager with the additional matching conditions of the delivery order, thereby making the allocation of the delivery order more relaxed and improving the allocation success rate, or making the allocation of the delivery order more strict and improving the delivery accuracy.

[0199] The capacity scheduler can summarize the matching results of all delivery components and capacity queues, and then judge whether the current allocation of the target project is successful based on the pre-set judgment conditions. For example, the pre-set judgment conditions can be: if all delivery components are successfully matched with the capacity queue, the current allocation of the target project is successful; or, if more than a certain proportion of delivery components are successfully matched with the capacity queue, the current allocation of the target project is successful.

[0200] The capacity scheduler can allocate the capacity unit to the matching capacity queue by recording the project information of the allocated target project, the size of the delivery component, and updating the traffic usage information in the management information of the capacity unit. When the allocation is completed, the capacity scheduler can also feedback to the user the information that the target project has been successfully created.

[0201] Continuing with the previous example, assume that the capacity scheduler obtains the remaining available delivery of the capacity queues corresponding to the three time periods from 5 to 8 o'clock from the capacity manager, which are assumed to be 12 million (capacity queue 1 corresponding to 5 to 6 o'clock), 14 million (capacity queue 2 corresponding to 6 to 7 o'clock), and 14 million (capacity queue 3 corresponding to 7 to 8 o'clock). The capacity scheduler determines that the delivery component of the time period corresponding to delivery order 1 is successfully matched based on the obtained remaining available delivery, and then records the order information in the management information of the corresponding capacity queue, where sub-order 1-1 and the corresponding delivery component of 6 million are recorded in the management information of capacity queue 1, sub-order 1-2 and the corresponding delivery component of 7 million are recorded in the management information of capacity queue 2, and sub-order 1-3 and the corresponding delivery component of 7 million are recorded in the management information of capacity queue 3, and the remaining available delivery is updated to 6 million (capacity queue 1), 7 million (capacity queue 2), and 7 million (capacity queue 3).

[0202] The capacity scheduler determines that the delivery order 1 is created successfully, updates the remaining usage quota of delivery user A to 180 million (the used quota is 20 million), and returns the updated remaining usage quota information to delivery user A. Based on this, at any subsequent time, assuming that the remaining usage quota of delivery user A is 10 million, if delivery user A creates another order of 20 million, the capacity scheduler will determine that 10 million is less than 20 million and refuse to create the order.

[0203] After the successful creation of delivery order 1, suppose that delivery user B creates delivery order 2 for 40 million delivery between 5:00 and 8:00 on the same day in the future. The total delivery quota of delivery user B is 500 million. The capacity scheduler determines that the remaining usage quota of 500 million is greater than 40 million, and delivery order 2 can be created. The capacity scheduler distributes the target delivery volume of 20 million to the above three time periods in the ratio of 1.2:1.4:1.4, and obtains delivery components of 12 million, 14 million and 14 million. The capacity scheduler matches the delivery components of 12 million, 14 million and 14 million with the remaining available delivery volume of 6 million (capacity queue 1), 7 million (capacity queue 2) and 7 million (capacity queue 3) of the corresponding capacity queues respectively. If delivery order 2 does not set a more relaxed additional matching condition, the matching result is failure, and delivery user B can be prompted that there is no remaining available delivery volume for delivery. Delivery user B can modify the order content of delivery order 2, or set a more relaxed additional matching condition for the current delivery order 2. Assuming that user B sets a delay of 5 hours to select a suitable delivery period for matching based on the original end time of delivery order 2, the capacity scheduler can find a suitable capacity queue for matching in the capacity queue within the range of 8:00-15:00. The matching success rate can be improved by more relaxed additional matching conditions. Assuming that after matching based on the additional matching conditions, the remaining available delivery volume of the capacity queue corresponding to the time period within the range of 8:00-11:00 successfully matches the delivery components of sub-order 2-1, sub-order 2-2, and sub-order 2-3, the capacity scheduler records the order content and corresponding delivery components of the matched sub-order in the management information of the capacity queue corresponding to 8:00-11:00, and updates the remaining available delivery volume of the capacity queue. The capacity manager can also record the order content of total order 2 and the delivery status information of the capacity queue. When the capacity scheduler successfully creates delivery order 2, it updates the remaining usage quota of delivery user B to 460 million (the used quota is 40 million), and returns the creation result to delivery user B.

[0204] In one or more embodiments, before allocating multiple delivery components to matching capacity queues, congestion prediction and / or delivery completion prediction can be performed on the capacity unit to be allocated, and the congestion prediction result and / or delivery completion prediction result of the capacity unit can be obtained, and the congestion prediction result and / or delivery completion prediction result can be returned to the user, so that the user can timely understand the completion status of the delivery order before the actual allocation and order completion creation.

[0205] Continuing with the previous example, after delivery user A and delivery user B create a delivery order, they can predict the congestion prediction result and the delivery completion prediction result based on the remaining available delivery volume and delivery component of the matching capacity queue, and send the congestion prediction result and the delivery completion prediction result to the delivery user, so that the delivery user can decide whether to modify the order content of the delivery order based on the congestion prediction result and the delivery completion prediction result.

[0206] Step 620: The capacity scheduler delivers the notes of the delivery component to the promotion position of the social platform within the delivery period of the allocated capacity queue.

[0207] For example, at any time during the delivery period corresponding to the capacity queue, the order information of the notes to be delivered can be obtained from the management information of the capacity queue, the notes can be delivered, and the delivery status information can be updated in a timely manner.

[0208] In this embodiment, the time range is divided into the smallest unit of time period, and a capacity queue corresponding to the time period is created to manage the traffic of the time period, so as to realize the adaptive management and scheduling of note delivery based on the capacity queue. When an order creation request arrives, the capacity queue is matched for the order, and the note delivery is performed when the target project is successfully allocated this time according to the matching result, or the creation request is rejected when the allocation fails. Specifically, in this embodiment, the target delivery amount of the delivery order is divided into time periods, and the delivery components are matched with the remaining available delivery amount of the corresponding capacity queue according to the time period, so that when the target project is successfully allocated this time according to the matching result, the delivery components are allocated to the corresponding capacity queue, and the note delivery is performed based on the capacity queue. In the case of a matching failure, the delivery order can be rejected. Since the note delivery is managed and scheduled by the capacity queue, and the maximum available delivery amount corresponding to the capacity queue is obtained by the traffic prediction model based on the traffic predicted by the time period corresponding to the capacity queue. Therefore, the maximum available delivery amount of the capacity queue is adaptively adjusted in different time periods according to the prediction results of the traffic prediction model, and adapted to the traffic difference between the peak and low peak periods of the platform. The notes are delivered on the premise that they can be completed, so that the completion rate of the note delivery in this embodiment is high and the delivery effect is good. In addition, this embodiment also records the management information of the capacity queue, realizing efficient management of the capacity queue and accurate delivery of notes. Furthermore, congestion prediction and order completion rate prediction are provided before the order is created, so that users can timely understand the congestion and / or completion status of the notes to be delivered, so as to choose whether to continue delivery, further improving the delivery completion rate and user experience.

[0209] Corresponding to the above method embodiment, this specification also provides a multimedia material scheduling device embodiment, Figure 8 FIG. 1 shows a schematic diagram of the structure of a multimedia material scheduling device provided by an embodiment of this specification. Figure 8As shown, the device comprises:

[0210] The project determination module 802 is configured to determine, in response to receiving a multimedia material delivery request, a target project corresponding to the multimedia material delivery request and a delivery time range and a target delivery amount corresponding to the target project;

[0211] The component determination module 804 is configured to determine a plurality of delivery time periods corresponding to the delivery time range according to a preset time period division rule, and determine the delivery components of the target delivery amount distributed to the plurality of delivery time periods;

[0212] The component matching module 806 is configured to match the delivery components corresponding to the multiple delivery periods respectively with the remaining available delivery amounts of the corresponding capacity units to obtain a matching result, wherein one capacity unit corresponds to one delivery period, the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, and the maximum available delivery amount is obtained based on the traffic predicted by the traffic prediction model for the corresponding delivery period;

[0213] The component allocation module 808 is configured to allocate the multiple delivery components to the matching capacity units when it is determined that the target project is successfully allocated this time according to the matching results of the delivery components corresponding to the multiple delivery time periods;

[0214] The component delivery module 810 is configured to deliver the multimedia material of the delivery component to the promotion position of the social platform within the delivery period of the allocated capacity unit.

[0215] Optionally, the device further comprises:

[0216] a forecast range determination module configured to determine a target time range;

[0217] A prediction period determination module is configured to determine a plurality of time periods to be measured corresponding to the target time range according to the preset time period division rule;

[0218] A capacity unit creation module is configured to create corresponding capacity units for the multiple time periods to be tested, respectively, wherein the capacity unit corresponding to the time period to be tested is consistent with the capacity unit corresponding to the delivery time period belonging to the same time period;

[0219] The traffic prediction module is configured to use the traffic prediction model to predict the traffic corresponding to the social platform in any time period to be tested, obtain the traffic prediction value corresponding to any time period to be tested, and determine the maximum available delivery volume of the capacity unit corresponding to any time period to be tested based on the traffic prediction value corresponding to any time period to be tested.

[0220] Optionally, the device further comprises:

[0221] The management module is configured to, after creating corresponding capacity units for the multiple time periods to be tested, respectively, record the time period to be tested information, maximum available delivery amount information and traffic usage information corresponding to the capacity unit in the management information of the capacity unit, wherein the traffic usage information includes remaining available delivery amount information and / or allocated delivery amount information.

[0222] The component allocation module is configured to record the project information of the allocated target project, the allocated delivery component size and the delivery status information of the multimedia material of the delivery component in the management information of the capacity unit, and update the traffic usage information according to the size of the delivery component.

[0223] Optionally, the component determination module is configured to divide the target delivery amount according to the ratio of the maximum available delivery amounts of the capacity units corresponding to the multiple delivery time periods, and determine the delivery components corresponding to the multiple delivery time periods.

[0224] Optionally, the component matching module is configured to obtain additional matching conditions of the target project, and match the delivery components corresponding to the multiple delivery time periods with the remaining available delivery quantities of the corresponding capacity units according to the additional matching conditions.

[0225] Optionally, the traffic prediction module is configured to input any time period to be tested and traffic influencing factor data into the traffic prediction model, and the traffic prediction model predicts the corresponding traffic prediction value of any time period to be tested under the influence of the traffic influencing factor data. The traffic prediction model is obtained based on the historical traffic data of the social platform, and the historical traffic data includes: multiple historical time periods belonging to the same time period as the any time period to be tested, historical traffic influencing factor data corresponding to at least one of the historical time periods, and the true traffic value of the social platform in the multiple historical time periods.

[0226] Optionally, the device further comprises:

[0227] A congestion estimation module is configured to obtain the allocated delivery amount of the capacity unit to which the delivery component is to be allocated, and perform congestion prediction and / or delivery completion prediction on the capacity unit according to the delivery component and the allocated delivery amount to obtain a congestion prediction result and / or delivery completion prediction result of the capacity unit;

[0228] The congestion information feedback module is configured to send the congestion prediction result and / or the delivery completion prediction result to the user of the target project.

[0229] Optionally, the device also includes: a delivery confirmation module, which is configured to receive delivery confirmation information returned by the user regarding the congestion prediction result and / or delivery completion prediction result before allocating the multiple delivery components to matching capacity units, and when it is determined according to the delivery confirmation information that the user confirms the execution of the delivery, the component allocation module enters the step of allocating the multiple delivery components to matching capacity units.

[0230] The above is a schematic scheme of a multimedia material scheduling device of this embodiment. It should be noted that the technical scheme of the multimedia material scheduling device and the technical scheme of the multimedia material scheduling method mentioned above belong to the same concept, and the details of the technical scheme of the multimedia material scheduling device that are not described in detail can all be referred to the description of the technical scheme of the multimedia material scheduling method mentioned above.

[0231] Corresponding to the above-mentioned note scheduling method embodiment of the social platform, this specification also provides a note scheduling device embodiment of the social platform, Fig. 9 FIG. 1 shows a schematic diagram of a note scheduling device for a social platform provided by an embodiment of the present specification. Fig. 9 As shown, the device comprises:

[0232] The note delivery response module 902 is configured to, in response to receiving a note delivery request, determine a target project corresponding to the note delivery request and a delivery time range and a target delivery amount corresponding to the target project;

[0233] The note delivery dispersing module 904 is configured to determine a plurality of delivery time periods corresponding to the delivery time range according to a preset time period division rule, and determine delivery components of the target delivery amount distributed to the plurality of delivery time periods;

[0234] The note component matching module 906 is configured to match the delivery components corresponding to the multiple delivery periods respectively with the remaining available delivery amounts of the corresponding capacity units to obtain a matching result, wherein one capacity unit corresponds to one delivery period, the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, and the maximum available delivery amount is obtained based on the traffic predicted by the traffic prediction model for the corresponding delivery period;

[0235] The note component allocation module 908 is configured to allocate the multiple delivery components to the matching capacity units when it is determined that the target project is successfully allocated this time according to the matching results of the delivery components corresponding to the multiple delivery time periods;

[0236] The note delivery module 910 is configured to deliver the notes of the delivery component to the promotion position of the social platform within the delivery period of the allocated capacity unit.

[0237] The above is a schematic scheme of a note scheduling device for a social platform of this embodiment. It should be noted that the technical scheme of the note scheduling device for the social platform and the technical scheme of the note scheduling method for the above-mentioned social platform belong to the same concept, and the details not described in detail in the technical scheme of the note scheduling device for the social platform can be found in the description of the technical scheme of the note scheduling method for the above-mentioned social platform.

[0238] Fig.10 The block diagram of a computing device 1000 according to an embodiment of the present specification is shown. The components of the computing device 1000 include but are not limited to a memory 1010 and a processor 1020. The processor 1020 is connected to the memory 1010 via a bus 1030, and the database 1050 is used to store data.

[0239] The computing device 1000 also includes an access device 1040 that enables the computing device 1000 to communicate via one or more networks 1060. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 1040 may include one or more of any type of network interface (e.g., a network interface card (NIC)) of wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a world-wide interoperability for microwave access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, and a near field communication (NFC).

[0240] In one embodiment of the present specification, the above components of the computing device 1000 and Fig.10 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Fig.10 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0241] The computing device 1000 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 1000 may also be a mobile or stationary server.

[0242] The memory 1010 is used to store computer programs / instructions, and the processor 1020 is used to execute the computer programs / instructions stored in the memory 1010. When the computer programs / instructions are executed by the processor, the steps of the method described in any of the above embodiments are implemented.

[0243] The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the method described in any of the above embodiments belong to the same concept, and the details not described in detail in the technical scheme of the computing device can be referred to the description of the technical scheme of the method described in any of the above embodiments.

[0244] An embodiment of the present specification further provides a computer-readable storage medium storing a computer program / instruction, which implements the steps of the method described in any of the above embodiments when executed by a processor.

[0245] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium belongs to the same concept as the technical scheme described in any of the above embodiments, and the details not described in detail in the technical scheme of the storage medium can be referred to the description of the technical scheme of the method described in any of the above embodiments.

[0246] An embodiment of the present specification also provides a computer program product, including a computer program / instruction, which implements the steps of the method described in any of the above embodiments when executed by a processor.

[0247] The above is a schematic scheme of a computer program product of this embodiment. It should be noted that the technical scheme of the computer program product and the technical scheme of the method described in any of the above embodiments belong to the same concept, and the details not described in detail in the technical scheme of the computer program product can be referred to the description of the technical scheme of the method described in any of the above embodiments.

[0248] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0249] The computer program / instruction includes computer program code, which may be in source code form, object code form, executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer readable medium may be appropriately increased or decreased according to the requirements of patent practice. For example, in some regions, according to patent practice, computer readable media do not include electric carrier signals and telecommunication signals.

[0250] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of this specification are not limited by the order of the actions described, because according to the embodiments of this specification, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of this specification.

[0251] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0252] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The optional embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that technicians in the relevant technical field can understand and use this specification well. This specification is limited only by the claims and their full scope and equivalents.

Claims

1. A multimedia material scheduling method, characterized in that: include: In response to receiving a multimedia material delivery request, determining a target project corresponding to the multimedia material delivery request and a delivery time range and a target delivery amount corresponding to the target project; According to a preset time period division rule, a plurality of delivery time periods corresponding to the delivery time range are determined, and delivery components of the target delivery amount distributed to the plurality of delivery time periods are determined; Matching the delivery components corresponding to the multiple delivery periods respectively with the remaining available delivery amounts of the corresponding capacity units to obtain a matching result, wherein one capacity unit corresponds to one delivery period, the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, and the maximum available delivery amount is obtained based on the traffic predicted by the traffic prediction model for the corresponding delivery period; When it is determined that the target project is successfully allocated this time according to the matching results of the delivery components respectively corresponding to the multiple delivery time periods, the multiple delivery components are allocated to the matching capacity units; The multimedia materials of the delivery component are delivered to the promotion position of the social platform within the delivery period of the allocated capacity unit.

2. The method according to claim 1, characterized in that Also includes: Determine target time frame; According to the preset time period division rule, determining a plurality of time periods to be tested corresponding to the target time range; Creating corresponding capacity units for the multiple time periods to be tested, respectively, wherein the capacity unit corresponding to the time period to be tested is consistent with the capacity unit corresponding to the delivery time period belonging to the same time period; Using the traffic prediction model, predict the traffic corresponding to any time period to be tested on the social platform to obtain a traffic prediction value corresponding to any time period to be tested; According to the traffic forecast value corresponding to any time period to be tested, the maximum available delivery volume of the capacity unit corresponding to any time period to be tested is determined.

3. The method according to claim 2, characterized in that After creating corresponding capacity units for the multiple time periods to be tested respectively, the method further includes: In the management information of the capacity unit, the information of the time period to be measured, the maximum available delivery amount information and the flow usage information corresponding to the capacity unit are recorded, and the flow usage information includes the remaining available delivery amount information and / or the allocated delivery amount information; The allocating the multiple delivery components to matching capacity units includes: In the management information of the capacity unit, the project information of the allocated target project, the allocated delivery component size, and the delivery status information of the multimedia material of the delivery component are recorded, and the traffic usage information is updated according to the size of the delivery component.

4. The method according to claim 1, characterized in that: Determining the target delivery amount to be distributed to delivery components of the plurality of delivery time periods includes: The target delivery amount is divided according to the ratio of the maximum available delivery amounts of the capacity units respectively corresponding to the multiple delivery time periods, so as to determine the delivery components respectively corresponding to the multiple delivery time periods.

5. The method according to claim 1, characterized in that Matching the delivery components corresponding to the multiple delivery time periods respectively with the remaining available delivery amounts of the corresponding capacity units includes: Get additional matching conditions for the target item; According to the additional matching condition, the delivery components corresponding to the multiple delivery time periods are matched with the remaining available delivery quantities of the corresponding capacity units.

6. The method according to claim 5, characterized in that The additional matching conditions include: any one or more of the following conditions: Matching conditions for matching capacity units corresponding to a specified target period; A matching condition for matching according to the priority information of the target project; The delivery period is adjusted according to the period adjustment strategy, and the matching conditions are matched according to the adjusted delivery period, the period adjustment strategy includes: adjusting the start time and end time of the target delivery period, and the duration of the target delivery period before and after the adjustment remains unchanged, and the target delivery period is any one or more delivery periods among the multiple delivery periods.

7. The method according to claim 2, characterized in that The traffic prediction model is used to predict the traffic of the social platform in any time period to be tested, and the traffic prediction value corresponding to any time period to be tested is obtained, including: Input any time period to be measured and flow influencing factor data into the flow prediction model, and use the flow prediction model to predict the corresponding flow prediction value of any time period to be measured under the influence of the flow influencing factor data; Among them, the traffic prediction model is obtained based on the historical traffic data of the social platform, and the historical traffic data includes: multiple historical time periods belonging to the same time period as any time period to be measured, historical traffic impact factor data corresponding to at least one of the historical time periods, and the true value of the traffic of the social platform in the multiple historical time periods.

8. The method according to claim 1, characterized in that Also includes: Obtaining the allocated delivery amount of the capacity unit to which the delivery component is to be allocated; According to the delivery component and the allocated delivery amount, congestion prediction and / or delivery completion prediction are performed on the capacity unit to obtain a congestion prediction result and / or delivery completion prediction result of the capacity unit; The congestion prediction result and / or delivery completion prediction result are sent to the user of the target project.

9. The method according to claim 8, characterized in that Before allocating the multiple delivery components to the matching capacity units, the method further includes: Receiving delivery confirmation information returned by the user in response to the congestion prediction result and / or delivery completion prediction result; When it is determined according to the delivery confirmation information that the user confirms to execute the delivery, the step of allocating the multiple delivery components to matching capacity units is entered.

10. A note scheduling method for a social platform, characterized in that: Applicable to servers, including: In response to receiving a note delivery request, determining a target project corresponding to the note delivery request and a delivery time range and a target delivery amount corresponding to the target project; According to a preset time period division rule, a plurality of delivery time periods corresponding to the delivery time range are determined, and delivery components of the target delivery amount distributed to the plurality of delivery time periods are determined; Matching the delivery components corresponding to the multiple delivery periods respectively with the remaining available delivery amounts of the corresponding capacity units to obtain a matching result, wherein one capacity unit corresponds to one delivery period, the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, and the maximum available delivery amount is obtained based on the traffic predicted by the traffic prediction model for the corresponding delivery period; When it is determined that the target project is successfully allocated this time according to the matching results of the delivery components respectively corresponding to the multiple delivery time periods, the multiple delivery components are allocated to the matching capacity units; The notes of the delivery component are delivered to the promotion position of the social platform within the delivery period of the allocated capacity unit.

11. A multimedia material dispatching device, characterized in that: include: A project determination module is configured to, in response to receiving a multimedia material delivery request, determine a target project corresponding to the multimedia material delivery request and a delivery time range and a target delivery amount corresponding to the target project; A component determination module is configured to determine a plurality of delivery time periods corresponding to the delivery time range according to a preset time period division rule, and determine delivery components of the target delivery amount distributed to the plurality of delivery time periods; A component matching module is configured to match the delivery components corresponding to the multiple delivery time periods with the remaining available delivery amounts of the corresponding capacity units to obtain a matching result, wherein one capacity unit corresponds to one delivery time period, the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, and the maximum available delivery amount is obtained based on the traffic predicted by the traffic prediction model for the corresponding delivery time period; A component allocation module is configured to allocate the multiple delivery components to matching capacity units when it is determined that the target project is successfully allocated this time according to the matching results of the delivery components corresponding to the multiple delivery time periods; The component delivery module is configured to deliver the multimedia materials of the delivery components to the promotion position of the social platform within the delivery period of the allocated capacity unit.

12. A note scheduling device for a social platform, characterized in that: include: A note delivery response module is configured to, in response to receiving a note delivery request, determine a target project corresponding to the note delivery request and a delivery time range and a target delivery amount corresponding to the target project; The note delivery dispersion module is configured to determine a plurality of delivery periods corresponding to the delivery time range according to a preset time period division rule, and determine delivery components of the target delivery amount dispersed to the plurality of delivery periods; The note component matching module is configured to match the delivery components corresponding to the multiple delivery periods respectively with the remaining available delivery amounts of the corresponding capacity units to obtain a matching result, wherein one capacity unit corresponds to one delivery period, the remaining available delivery amount is determined according to the maximum available delivery amount and the allocated delivery amount of the capacity unit to which it belongs, and the maximum available delivery amount is obtained based on the traffic predicted by the traffic prediction model for the corresponding delivery period; The note component allocation module is configured to allocate the multiple delivery components to the matching capacity units when it is determined that the target project is successfully allocated this time according to the matching results of the delivery components corresponding to the multiple delivery time periods; The note delivery module is configured to deliver the notes of the delivery component to the promotion position of the social platform within the delivery period of the allocated capacity unit.

13. A computing device, characterized in that: include: Memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the computer program / instructions are executed by the processor, the steps of the method described in any one of claims 1 to 9 are implemented.

14. A computer-readable storage medium, characterized in that: It stores a computer program / instruction, which implements the steps of the method described in any one of claims 1 to 9 when executed by a processor.

15. A computer program product, characterized in that The method comprises a computer program / instruction which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 9.

Images