Advertisement flow distribution method, device, equipment and storage medium

Abstract

The present invention relates to the field of Internet technology, and discloses an advertising traffic distribution method, device, equipment and storage medium. The advertising traffic distribution method comprises: receiving an advertising request initiated by a client, wherein the advertising request contains resource position information; based on the resource position information, constructing a traffic distribution model corresponding to the current business scenario, the traffic distribution model includes request components of several advertising channels and corresponding traffic distribution links; based on the traffic distribution model, the advertising request flows into the request component of the corresponding advertising channel according to the traffic distribution link for advertisement filling. The present invention realizes the refinement of the advertising traffic distribution strategy to the resource position dimension, and then can perform full-link traffic distribution modeling according to specific business scenarios, while ensuring the diversity of traffic sales methods, and improving the flexibility of traffic distribution by dynamically configuring traffic distribution links.

Description

Technical Field

[0001] The present invention relates to the field of Internet technology, and in particular to an advertisement traffic distribution method, device, equipment and storage medium. Background Art

[0002] In the Internet digital age, advertising monetization is one of the main ways for applications to make profits by displaying advertisements. It is also a reliable and healthy business model for enterprises to monetize traffic. The traffic allocation algorithm is one of the core factors affecting advertising revenue. Currently, the advertising traffic allocation algorithm is mainly divided into two modes:

[0003] (1) The waterfall flow model refers to distributing advertising traffic to different advertising channels through the aggregation platform. Each advertising channel makes requests from top to bottom according to the weight. The advertising channels with high weights first decide whether to purchase advertising traffic, while those with low weights need to wait for the distribution of the remaining traffic.

[0004] (2) Head bidding mode means that a price inquiry is initiated to each advertising channel at the same time, and the advertising traffic is sold to the advertising channel with the highest bid.

[0005] However, the above two advertising traffic allocation algorithms each have their own shortcomings. For example, in the waterfall mode, as the number of advertising channels continues to increase, the maintenance cost of the traffic allocation link will increase, the request time will increase, and the CPM (Cost Per Mille, cost per thousand impressions) is not optimal. In the header bidding mode, the advertising sales method is single, and it is impossible to provide price or volume guarantee for a certain advertising channel. In addition, in different applications and in different resource advertising positions, the advertising realization behavior and efficiency are not completely consistent, so it is necessary to adjust the traffic allocation according to the specific business scenario, and the traditional waterfall mode and header bidding mode cannot effectively solve this problem. Summary of the invention

[0006] The main purpose of the present invention is to provide an advertising traffic distribution method, device, equipment and storage medium, aiming to solve the technical problem of how to adapt to the advertising traffic distribution of different business scenarios and ensure the diversity of advertising sales methods and low link consumption.

[0007] A first aspect of the present invention provides an advertisement traffic allocation method, the advertisement traffic allocation method comprising:

[0008] Receiving an advertisement request initiated by a client, wherein the advertisement request includes resource location information;

[0009] Based on the resource location information, a traffic allocation model corresponding to the current business scenario is constructed, wherein the traffic allocation model includes request components of several advertising channels and corresponding traffic allocation links;

[0010] Based on the traffic distribution model, the advertisement request is flowed into the request component of the corresponding advertisement channel according to the traffic distribution link for advertisement filling.

[0011] Optionally, in a first implementation manner of the first aspect of the present invention, constructing a traffic allocation model corresponding to the current business scenario based on the resource location information includes:

[0012] Based on the resource position information, determine the business scenario corresponding to the current advertising traffic allocation;

[0013] Configure the traffic distribution strategy corresponding to the business scenario;

[0014] Based on the traffic allocation strategy, a traffic allocation model corresponding to the current business scenario is constructed.

[0015] Optionally, in a second implementation of the first aspect of the present invention, constructing a traffic distribution model corresponding to the current business scenario based on the traffic distribution strategy includes:

[0016] Based on the traffic distribution strategy, determining an assembly mode of request components of each advertising channel;

[0017] Based on the assembly mode, the request components of each advertising channel are dynamically assembled to obtain the traffic allocation link corresponding to each advertising channel, so as to form a traffic allocation model corresponding to the current business scenario.

[0018] Optionally, in a third implementation of the first aspect of the present invention, the assembly mode includes a waterfall assembly mode and a hierarchical flattening assembly mode;

[0019] The types of the request components include a first request component that only supports the waterfall assembly mode and a second request component that supports both the waterfall assembly mode and the level-flattening assembly mode.

[0020] Optionally, in a fourth implementation of the first aspect of the present invention, the traffic allocation model includes a waterfall allocation model, a real-time price comparison allocation model, and a hybrid allocation model;

[0021] Among them, each request component in the waterfall flow allocation model adopts the waterfall flow mode for traffic allocation; each request component in the real-time price comparison allocation model adopts the header bidding mode for traffic allocation; each request component in the hybrid allocation model mainly uses the waterfall flow mode for traffic allocation, and partially adopts the header bidding mode for traffic allocation.

[0022] Optionally, in a fifth implementation of the first aspect of the present invention, based on the traffic distribution model, flowing the advertisement request into a request component of a corresponding advertisement channel according to the traffic distribution link for advertisement filling includes:

[0023] When the traffic allocation model is the waterfall allocation model, the advertisement request is allocated according to the traffic allocation link corresponding to each request component. If the current request component responds to the advertisement request and fills the corresponding resource position with advertisement resources, the filled advertisement resources are returned to the client for advertisement delivery, and the remaining traffic of the advertisement request flows into the request component of the next priority advertisement channel to continue advertisement filling, wherein different advertisement channels correspond to different traffic allocation priorities;

[0024] When the traffic allocation model is the real-time price comparison allocation model, the advertising request is allocated according to the traffic allocation link corresponding to each request component. If there is currently a request component with the highest bid at the same priority level that responds to the advertising request and fills the corresponding resource position with advertising resources, the filled advertising resources are returned to the client for advertising delivery, wherein all advertising channels correspond to the same traffic allocation priority.

[0025] Optionally, in a sixth implementation of the first aspect of the present invention, the step of flowing the advertisement request into a request component of a corresponding advertisement channel according to the traffic distribution link based on the traffic distribution model to perform advertisement filling further includes:

[0026] When the traffic distribution model is the hybrid distribution model, the advertisement request is distributed according to the traffic distribution link corresponding to each request component;

[0027] If the type of the current request component is the first request component and it responds to the advertisement request and fills the corresponding resource position with advertisement resources, the filled advertisement resources are returned to the client for advertisement delivery, and the remaining traffic of the advertisement request is flowed into the request component of the next priority advertisement channel to continue advertisement filling;

[0028] If the type of the current request component is the second request component and the bid is the highest among those with the same priority, and the current request component responds to the advertising request and fills the corresponding resource positions with advertising resources, the filled advertising resources are returned to the client for advertising delivery, and the remaining traffic of the advertising request is flowed into the request component of the next priority advertising channel to continue advertising filling.

[0029] A second aspect of the present invention further provides an advertisement flow distribution device, the advertisement flow distribution device comprising:

[0030] A traffic receiving module, configured to receive an advertisement request initiated by a client, wherein the advertisement request includes resource location information;

[0031] A model building module, used to build a traffic allocation model corresponding to the current business scenario based on the resource position information, wherein the traffic allocation model includes request components of several advertising channels and corresponding traffic allocation links;

[0032] The traffic distribution module is used to flow the advertisement request into the request component of the corresponding advertisement channel according to the traffic distribution link for advertisement filling based on the traffic distribution model.

[0033] Optionally, in a first implementation of the second aspect of the present invention, the model building module includes:

[0034] A determination unit, configured to determine a business scenario corresponding to the current advertisement traffic allocation based on the resource position information;

[0035] A configuration unit, used to configure a traffic distribution strategy corresponding to the business scenario;

[0036] The construction unit is used to construct a traffic distribution model corresponding to the current business scenario based on the traffic distribution strategy.

[0037] Optionally, in a second implementation of the second aspect of the present invention, the construction unit is specifically used for:

[0038] Based on the traffic distribution strategy, determining an assembly mode of request components of each advertising channel;

[0039] Based on the assembly mode, the request components of each advertising channel are dynamically assembled to obtain the traffic allocation link corresponding to each advertising channel, so as to form a traffic allocation model corresponding to the current business scenario.

[0040] Optionally, in a third implementation of the second aspect of the present invention, the assembly mode includes a waterfall assembly mode and a hierarchical flattening assembly mode;

[0041] The types of the request components include a first request component that only supports the waterfall assembly mode and a second request component that supports both the waterfall assembly mode and the level-flattening assembly mode.

[0042] Optionally, in a fourth implementation of the second aspect of the present invention, the traffic allocation model includes a waterfall allocation model, a real-time price comparison allocation model, and a hybrid allocation model;

[0043] Among them, each request component in the waterfall flow allocation model adopts the waterfall flow mode for traffic allocation; each request component in the real-time price comparison allocation model adopts the header bidding mode for traffic allocation; each request component in the hybrid allocation model mainly uses the waterfall flow mode for traffic allocation, and partially adopts the header bidding mode for traffic allocation.

[0044] Optionally, in a fifth implementation of the second aspect of the present invention, the flow distribution module is specifically used to:

[0045] When the traffic allocation model is the waterfall allocation model, the advertisement request is allocated according to the traffic allocation link corresponding to each request component. If the current request component responds to the advertisement request and fills the corresponding resource position with advertisement resources, the filled advertisement resources are returned to the client for advertisement delivery, and the remaining traffic of the advertisement request flows into the request component of the next priority advertisement channel to continue advertisement filling, wherein different advertisement channels correspond to different traffic allocation priorities;

[0046] When the traffic allocation model is the real-time price comparison allocation model, the advertising request is allocated according to the traffic allocation link corresponding to each request component. If there is currently a request component with the highest bid at the same priority level that responds to the advertising request and fills the corresponding resource position with advertising resources, the filled advertising resources are returned to the client for advertising delivery, wherein all advertising channels correspond to the same traffic allocation priority.

[0047] Optionally, in a sixth implementation of the second aspect of the present invention, the flow distribution module is further used to:

[0048] When the traffic distribution model is the hybrid distribution model, the advertisement request is distributed according to the traffic distribution link corresponding to each request component;

[0049] If the type of the current request component is the first request component and it responds to the advertisement request and fills the corresponding resource position with advertisement resources, the filled advertisement resources are returned to the client for advertisement delivery, and the remaining traffic of the advertisement request is flowed into the request component of the next priority advertisement channel to continue advertisement filling;

[0050] If the type of the current request component is the second request component and the bid is the highest among those with the same priority, and the current request component responds to the advertising request and fills the corresponding resource positions with advertising resources, the filled advertising resources are returned to the client for advertising delivery, and the remaining traffic of the advertising request is flowed into the request component of the next priority advertising channel to continue advertising filling.

[0051] A third aspect of the present invention provides a computer device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the computer device executes the above-mentioned advertising traffic distribution method.

[0052] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned advertising traffic distribution method.

[0053] In the technical solution provided by the present invention, when allocating the advertising traffic of the client, firstly, according to the resource position information contained in the advertising request, a traffic allocation model including request components of several advertising channels and corresponding traffic allocation links corresponding to the current business scenario (corresponding to the advertising resource position of the current client) is constructed, and then the advertising request (that is, the advertising traffic) is flowed into the request components of different advertising channels according to the traffic allocation links set in the traffic allocation model corresponding to the constructed current business scenario, and the advertisements are filled in sequence, thereby realizing the advertising traffic allocation. The present invention realizes the refinement of the advertising traffic allocation strategy to the resource position dimension, and then the full-link traffic allocation model can be performed according to the specific business scenario, while ensuring the diversity of traffic sales methods, and the flexibility of traffic allocation is improved by dynamically configuring the traffic allocation link. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A schematic diagram of an embodiment of the advertising traffic allocation method in an embodiment of the present invention;

[0055] Figure 2 It is a schematic diagram of advertisement flow distribution corresponding to the waterfall flow distribution model in an embodiment of the present invention;

[0056] Figure 3 This is a schematic diagram of advertising traffic allocation corresponding to the real-time price comparison allocation model in an embodiment of the present invention;

[0057] Figure 4 This is a schematic diagram of advertisement traffic allocation corresponding to the hybrid allocation model in an embodiment of the present invention;

[0058] Figure 5 A schematic diagram of an embodiment of an advertisement flow distribution device in an embodiment of the present invention;

[0059] Figure 6 FIG. 1 is a schematic diagram of an embodiment of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0061] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 , an embodiment of the advertising traffic allocation method in the embodiment of the present invention includes:

[0062] 101. Receive an advertisement request initiated by a client, wherein the advertisement request includes resource location information;

[0063] In this embodiment, the client is relative to the backend advertising system, which can be a mobile terminal, such as a mobile phone, tablet computer, or a computer. When the user uses the APP, application or web page installed on the client, the client automatically triggers an advertising request to the backend advertising system.

[0064] In this embodiment, the advertisement request includes resource position information, which specifically refers to the position of the advertisement in the APP, application or webpage on the current client, that is, the specific position of the advertisement to be displayed on the client, such as the top banner of the homepage, the middle interstitial of the article page, etc. The resource position information mainly refers to the advertisement delivery strategy information, including traffic allocation method (waterfall flow / head bidding), advertisement position frequency control, advertisement position targeting, advertisement position base price, material size / template, etc.

[0065] 102. Based on the resource position information, construct a traffic allocation model corresponding to the current business scenario, wherein the traffic allocation model includes request components of several advertising channels and corresponding traffic allocation links;

[0066] In this embodiment, based on the resource location information carried in the advertising request, the advertising system will build a traffic allocation model for the current business scenario. The model includes request components of multiple advertising channels and traffic allocation links corresponding to these advertising channels. Each advertising channel represents an advertiser, advertising network or advertising source, and the request component is a specific module for processing advertising requests. The traffic allocation link is the path of the advertising request starting from the receiving point, through a series of decisions and processing, and finally reaching the request component corresponding to a certain advertising channel. The design of the link takes into account many factors, such as the quality of the advertisement, the advertiser's bid, the priority of the advertising channel, etc.

[0067] In an optional embodiment, the above step 102 further includes:

[0068] 1021. Determine a business scenario corresponding to the current advertisement traffic allocation based on the resource position information;

[0069] Different resource positions correspond to different display locations of ads on web pages, applications or other media. Different resource positions may correspond to different user groups, access behaviors or advertising effects. Therefore, based on resource position information, such as traffic allocation method (waterfall flow / head bidding), ad position frequency control, ad position targeting, ad position floor price, material size / template, etc., the business scenario corresponding to the current advertising traffic allocation can be determined, that is, the specific business environment or context in which the advertising traffic allocation is located. For example, a resource position at the top of the homepage may correspond to a business scenario with high exposure and high click-through rate, while a resource position at the bottom of the content page may correspond to a business scenario with lower exposure but more accurate user reach.

[0070] 1022. Configure a traffic distribution strategy corresponding to the business scenario;

[0071] In this embodiment, in order to achieve flexible allocation of advertising traffic in different business scenarios, it is necessary to pre-configure the traffic allocation strategy corresponding to the business scenario. The traffic allocation strategy is used to build a traffic allocation model. The configuration of the traffic allocation strategy includes considerations based on multiple factors such as user portraits, advertising content, advertising effect prediction (such as click-through rate, conversion rate), and bidding of advertising channels. It supports manual configuration and program automatic configuration (such as configuration through artificial intelligence large models), specifically including configuring the allocation weights of different resource positions, the traffic allocation priority of each advertising channel, and the advertising traffic allocation algorithm supported by the request component of each advertising channel (waterfall mode and / or header bidding mode).

[0072] 1023. Based on the traffic allocation strategy, construct a traffic allocation model corresponding to the current business scenario.

[0073] In this embodiment, the traffic allocation model is a mathematical model or algorithm that concretizes the advertising traffic allocation strategy, which is used to realize the assembly of the request components of each advertising channel and form a traffic allocation link. The traffic allocation model can dynamically adjust the traffic allocation link according to different traffic allocation strategies, and then adjust the advertising display of the client, thereby maximizing the advertising effect or meeting other business goals.

[0074] In an optional embodiment, the above step 1023 further includes:

[0075] (1) determining an assembly mode of request components of each advertising channel based on the traffic distribution strategy;

[0076] (2) Based on the assembly mode, dynamically assemble the request components of each advertising channel to obtain the traffic allocation link corresponding to each advertising channel to form a traffic allocation model corresponding to the current business scenario.

[0077] The traffic allocation strategy is used to determine how to allocate advertising traffic to different advertising channels, including the allocation weights of different resource positions, the traffic allocation priority of each advertising channel, the advertising traffic allocation algorithm (waterfall mode and / or header bidding mode) supported by the request component of each advertising channel, etc. This embodiment supports dynamically adjusting the traffic allocation strategy according to the advertising business characteristics of specific resource positions, that is, adjusting the priority of the request component of each advertising channel, and then adjusting the priority of each advertising channel.

[0078] The request component is a specific module for processing advertising requests. It uses the modeling concept to abstract and encapsulate the access logic of each advertising channel, extract common methods and encapsulate them into component form. The priority adjustment of the advertising link is essentially the priority adjustment of each request component, that is, the assembly of components can reflect the traffic distribution link of the entire advertising system.

[0079] The assembly mode refers to how to combine the request components of each advertising channel to form a complete advertising request processing flow. The assembly mode may vary from advertising channel to advertising channel. The corresponding request components of different advertising channels may adopt the same or different assembly modes. The hierarchy of the request components of each advertising channel and the order between the hierarchies are determined according to the traffic distribution strategy.

[0080] According to the assembly mode, the request components of each advertising channel are assembled according to the predetermined order and logic. Dynamic assembly means that this process can be performed at runtime, rather than statically hard-coded during development. By assembling the request components of each advertising channel, a complete traffic distribution link is formed from receiving the ad request to filling the ad content, ensuring that the ad request can be correctly processed and displayed to the user. Integrating the traffic distribution links of all advertising channels together forms a complete traffic distribution model. This traffic distribution model can dynamically allocate ad requests to different advertising channels according to the traffic distribution strategy, and process these requests to display ads on the client.

[0081] The traffic allocation model is built based on the needs and traffic allocation strategies of the current business scenarios. Therefore, it can flexibly adapt to changes in business scenarios, such as the addition of new advertising channels, changes in user portraits, etc., and can efficiently process advertising requests to ensure that advertisements can be displayed to users in the most appropriate way, thereby improving advertising effectiveness and user experience.

[0082] In an optional embodiment, the assembly mode includes:

[0083] (1) Waterfall assembly mode

[0084] The waterfall assembly mode refers to the use of waterfall flow to assemble request components of various advertising channels. Multiple levels with different traffic allocation priorities are formed between different request components. The traffic allocation priority of the request component at the previous level is greater than the traffic allocation priority of the request component at the next level.

[0085] (2) Hierarchical flattening assembly mode

[0086] The hierarchical flattening assembly mode refers to assembling the request components of each advertising channel in parallel. Different request components are at the same level, that is, the traffic allocation priority of each request component is the same, and there is no order of traffic allocation between request components.

[0087] In an optional embodiment, the types of the request components include a first request component that only supports the waterfall assembly mode and a second request component that supports both the waterfall assembly mode and the level-flattening assembly mode. The first request component only supports the waterfall mode for advertising traffic allocation, while the second request component supports both the waterfall mode and the header bidding mode for advertising traffic allocation.

[0088] In an optional embodiment, the traffic distribution model includes:

[0089] (1) Waterfall flow distribution model: In the waterfall flow distribution model, each request component uses the waterfall flow mode to distribute traffic.

[0090] For example, when advertising traffic is allocated to the first-level request component A, if request component A is not filled with ads or the filled ads do not meet the display conditions (such as the ad content does not meet the regulations, etc.), the advertising traffic will continue to flow to the next-level request component B. The traffic allocation process will continue until a request component at a certain level is filled with displayable ads. Once a request component at a certain level is filled with suitable ads, the advertising system will immediately return the ad to the client to display it to the user, and at the same time flow the remaining advertising traffic to the next-level request component. If the request components at all levels are not filled with ads, the client may display a default ad or its own product / service ad.

[0091] (2) Real-time price comparison allocation model: In the real-time price comparison allocation model, each request component uses the header bidding mode to allocate traffic.

[0092] For example, when advertising traffic is allocated to all request components at the same level at the same time, all request components use the header bidding model to allocate traffic. If request component A bids the highest and fills the ad, the advertising system will immediately return the ad to the client to display to the user. If all request components do not fill the ad, the client may display a default ad or its own product / service ad.

[0093] (3) Hybrid allocation model: In the hybrid allocation model, each request component mainly uses the waterfall flow mode to allocate traffic, and partially uses the header bidding mode to allocate traffic. The hybrid allocation model integrates the advantages of the waterfall flow allocation model and the real-time price comparison allocation model. It can not only ensure the diversity of advertising sales methods when the number of advertising channels increases, but also reduce the time consumption of the advertising traffic allocation link.

[0094] For example, when advertising traffic is allocated to the first-level request component A, if request component A is not filled with ads or the filled ads do not meet the display conditions (such as the ad content does not meet the regulations, etc.), the advertising traffic will continue to flow to the second-level request components B, C, and D, which use the header bidding model to allocate traffic. If request component B bids the highest and fills the ad, the advertising system will immediately return the ad to the client to display to the user, and at the same time, the remaining advertising traffic will flow to the next-level request component.

[0095] It should be noted that, although ideally there will be a corresponding advertising channel bidding for each advertising flow, and the winning bid will be returned to the client. However, in reality, there are cases of unsold goods, just like in the auction market. At this time, the traffic can only flow into the subsequent links through the subsequent waterfall flow or the configured traffic allocation strategy. There are many reasons for unsold goods due to the failure to fill in the advertisements in the price comparison link, such as: the price is lower than the configured minimum price, and it would rather fail than sell the traffic at a low price; the material is not compliant; the advertising channel decides not to purchase this traffic based on the user portrait; the material size and the actual display position cannot be adapted, etc.

[0096] 103. Based on the traffic distribution model, the advertisement request is flowed into the request component of the corresponding advertisement channel according to the traffic distribution link for advertisement filling.

[0097] In this embodiment, the advertising system will direct the advertising request to the request component corresponding to the corresponding advertising channel according to the link information in the traffic distribution model, and the request component of each advertising channel will fill the advertising request with appropriate advertising content according to its own situation. Finally, after traffic distribution and advertising filling, the client will receive a response filled with advertising content, which is selected from each advertising channel according to resource position information and current business scenarios.

[0098] In this embodiment, when allocating the advertising traffic of the client, firstly, according to the resource position information contained in the advertising request, a traffic allocation model including request components of several advertising channels and corresponding traffic allocation links corresponding to the current business scenario (corresponding to the advertising resource position of the current client) is constructed, and then the advertising request (i.e., advertising traffic) is flowed into the request components of different advertising channels according to the traffic allocation links set in the traffic allocation model corresponding to the constructed current business scenario, and the advertisements are filled in turn, thereby realizing the advertising traffic allocation. This embodiment realizes the refinement of the advertising traffic allocation strategy to the resource position dimension, and then the full-link traffic allocation model can be performed according to the specific business scenario, while ensuring the diversity of traffic sales methods, and improving the flexibility of traffic allocation by dynamically configuring the traffic allocation link.

[0099] To facilitate a better understanding of the present invention, the following provides examples of advertisement traffic allocation based on different traffic allocation models.

[0100] See also Figure 2 , Figure 2The following is a schematic diagram of advertisement traffic allocation corresponding to the waterfall allocation model in this embodiment. In this embodiment, when the traffic allocation model is the waterfall allocation model, advertisement requests are allocated according to the traffic allocation links corresponding to each request component. If the current request component responds to the advertisement request and fills the corresponding resource position with advertisement resources, the filled advertisement resources are returned to the client for advertisement delivery, and the remaining traffic of the advertisement request flows into the request component of the next priority advertisement channel to continue advertisement filling, wherein different advertisement channels correspond to different traffic allocation priorities.

[0101] like Figure 2 As shown, the waterfall distribution model includes request components of each advertising channel (request components 1-4) and traffic distribution links corresponding to each advertising channel (request component 1->request component 2->request component 3->request component 4). Assuming that there are three ad slots in the ad traffic (such as ad slots A, B, and C), the ad traffic is first allocated to the highest priority request component 1. If request component 1 fills ad slot A, that is, it purchases part of the ad traffic, the ad system returns the ad resources filled by request component 1 to the client for ad delivery, and allocates the remaining ad traffic (including ad slots B and C) to the next priority request component 2. If request component 2 does not fill ad slots B and C, that is, it does not purchase any ad traffic, the ad system allocates the remaining ad traffic (including ad slots B and C) to the next priority request component 3. If request component 3 also does not fill ad slots B and C, the ad system allocates the remaining ad traffic (including ad slots B and C) to the next priority request component 4. If request component 4 fills ad slots B and C, that is, it purchases the remaining ad traffic, the ad system returns the ad resources filled by request component 4 to the client for ad delivery.

[0102] like Figure 2As shown, in the waterfall flow allocation model, different advertising channels have different priorities, and one priority corresponds to one advertising channel. The priority is actually to show that the upper layer has a higher priority than the lower layer in the previous waterfall flow allocation model. One priority of the waterfall flow allocation model will only be given to one advertising channel. In the traditional waterfall-based traffic distribution link, the traffic first flows into the highest-level advertising channel (the highest priority channel). If the highest-level advertising channel purchases traffic, the advertising traffic will be cut off and the corresponding advertisement will be returned. If not, it will flow into the advertising channel of the next level, just like a waterfall, and the traffic is consumed layer by layer and decreases layer by layer. If the traffic distribution strategy is changed or a new advertising channel is connected, it is often necessary to develop, upgrade and redeploy the advertising system. Based on this, this embodiment realizes the decoupling between layers, that is, there is no longer a higher priority of level 1 than level 2, but each level is encapsulated as a component, each component supports the configuration of dynamic priority, and finally dynamically assembled into this traffic distribution link. Link disassembly is to decouple and develop the previous solidified links, so as to facilitate the subsequent abstraction into general components. Tier is a term in the waterfall allocation model, which represents the decrease in traffic after each channel is filled, or a description of priority. Channels are actually advertising channels and traffic buyers. Filled and unfilled actually refer to whether the buyers of this traffic at this level have purchased (filled ads). If they have purchased and filled ads, the link will go directly to the dotted part of the link to the ad return. If it is not filled, the ad traffic will continue to flow to the next level and ask the advertising channel at that level whether to purchase.

[0103] The traditional advertising system's waterfall-based traffic distribution link is fixed. The reason is that each advertising channel has different program processing logic, so the traffic distribution link can only be from advertising channel 1 to advertising channel 2, then to advertising channel 3, and so on. That is, the advertising system first asks the highest priority buyer (advertising channel 1) whether to purchase this traffic (fill ads). If not, it will ask the next priority buyer (advertising channel 2) again, thereby implementing serial traffic distribution logic. In the past, the advertising system was based on waterfall flow, and advertising channel 2 could only purchase the tail volume that advertising channel 1 did not purchase, which made the scalability and flexibility of the advertising link relatively low.

[0104] Different advertising channels provide different access methods or require different parameters for accessing APIs, but they also have commonalities. After the ads are filled, they are sent to end users. There are also commonalities, such as ads are nothing more than materials, including titles, pictures, videos, descriptions, etc. The same parameters of the requested ads, such as user portraits such as user age and gender, and the common parts of the ads returned by the interface are extracted. This is the commonality of attributes. Then, when traffic flows to each advertising channel, it must ask through the interface whether to make this purchase and how much to bid, and finally fill the ads. The subsequent links programmatically process the materials (labeling / review / adaptation, etc.) and have common behaviors, which can be abstracted into general methods, that is, the commonality of attributes and behaviors are abstracted and encapsulated into components. The waterfall distribution model of this embodiment decomposes the traffic distribution link based on the level / channel, so that the links of each level / channel are decoupled, and the modeling concept is used to abstract and encapsulate the access logic of each channel, extract the commonalities of functions and attributes, and encapsulate the access of each level / channel into a common component, and according to whether the header bidding mode is supported, the request component of each advertising channel can be obtained by inheritance. The priority adjustment of the advertising link in this embodiment is essentially the priority adjustment of each request component, that is, the assembly of components can be used to reflect what the traffic distribution link of the entire advertising system is like. This embodiment refines the advertising distribution link into resource bit dimensions, and through the abstraction and transformation of the link, the original solidified link can be dynamically configured through components, thereby solving the problem of fixed priority of each advertising channel.

[0105] In addition, it should be further explained that although some common attributes of most advertising channels on the market and some common behaviors of advertising links can be abstracted and encapsulated, the non-common parts need to be extended after inheriting the general components. For example, some advertising channels require some customized data or customized capabilities. After the general components are abstracted, the request components corresponding to each advertising channel can be generated, and then the request components can be assembled into traffic distribution links. The traffic distribution links are assembled according to the specific configuration of resource positions (configured by business products or operations according to contract orders or scenarios). Each request component has an attribute field to determine the hierarchical order between each advertising channel (that is, the priority of traffic distribution). All request components that inherit the general components support the waterfall flow level, and the request components that inherit the general components and adopt the header bidding mode not only support the waterfall flow level, but also support local real-time price comparison to achieve leveling. As long as the request component inherits the general component, it can be used to dynamically build a traffic distribution model.

[0106] See also Figure 3 , Figure 3This is a schematic diagram of advertising traffic allocation corresponding to the real-time price comparison allocation model in this embodiment. In this embodiment, when the traffic allocation model is the real-time price comparison allocation model, the advertising request is allocated according to the traffic allocation link corresponding to each request component. If there is currently a request component with the highest bid in the same priority level that responds to the advertising request and fills the corresponding resource position with advertising resources, the filled advertising resources are returned to the client for advertising delivery, wherein all advertising channels correspond to the same traffic allocation priority.

[0107] like Figure 3 As shown, the real-time price comparison allocation model includes the request components (request components 1-4) of each advertising channel and the traffic allocation links corresponding to each advertising channel (flowing to request components 1-4 at the same time). Assuming that there are three ad slots in the advertising traffic (such as ad slots A, B, and C), the advertising traffic is allocated to request components 1-4 at the same time, that is, request components 1-4 use the header bidding method to bid for ad slots A, B, and C at the same time. If request component 1 bids the highest for ad slot A, and request component 4 bids the highest for ad slots B and C, then request component 1 fills ad slot A, and request component 4 fills ad slots B and C, and then the advertising resources filled by request component 1 and request component 4 are returned to the client for advertising delivery.

[0108] In this embodiment, it is assumed that the flow distribution link based on waterfall flow is: first access advertising channel 1, and then access advertising channel 2, then the flow first flows into advertising channel 1, which takes time t1, and the flow that advertising channel 1 does not purchase flows to advertising channel 2, which takes time t2, then the total flow time of the flow distribution link is time t1 + time t2, that is, the more levels in the flow distribution link, the greater the total time, and the flow decreases layer by layer. If advertising channel 1 and advertising channel 2 adopt the local real-time bidding mode, the flow will flow into advertising channel 1 and advertising channel 2 at the same time, that is, the bids of advertising channels 1 and 2 are asked at the same time, and the advertising system selects the highest price as the seller's transaction, and sends the corresponding advertisement. The total link time is the maximum of time t1 and time t2. In other words, from the perspective of traffic distribution, advertising channel 1 and advertising channel 2 are channels of the same level. Because they receive the same advertising traffic, the level is flattened from 2 levels to 1 level, and the time consumption is changed from the sum of the time consumption of the two levels to the maximum time consumption of one level. In other words, the more advertising channels there are, the higher the efficiency of flattening the levels, and the link time consumption will be significantly reduced.

[0109] In this embodiment, the request component that supports the header bidding model can make parallel requests according to the business scenario to reduce the serialization of advertising channels, reduce the time consumption of link requests, and support real-time bidding to improve traffic monetization capabilities. In addition, the real-time price comparison distribution model also introduces distributed caching and asynchronous request solutions to solve the problem of how the SDK advertising access method adapts to the header bidding model. The specific implementation method is: the client requests the advertising system, generates a unique identifier RID (Request Identification), asynchronously and in parallel requests other header bidding request components, and stores them in the distributed cache based on the RID. In addition, the main thread sends a recall request to the client to recall the advertisement. The subsequent client carries the SDK advertisement into the traffic distribution link when making a second request. Here, the SDK access method refers to the advertising access method, because the component is abstracted by the common attributes and common behaviors of various advertising channels, but different advertising channels may provide different access methods, such as Figure 3 Request components 1-3 use the API access method, and request component 4 uses the SDK access method. In order to make the traffic distribution strategy flexible and compatible with most advertising channel providers in the market, this embodiment further designs the SDK access method based on the current mainstream API access method. It should be noted that SDK access requires user devices (such as mobile phones or tablets) to obtain advertising data, so part of the access code will be placed on the application software used by the user, and then the corresponding advertising data can be obtained by invoking SDK access.

[0110] In this embodiment, the request component that supports header bidding is compatible with the SDK access method when the levels are flattened. When traffic accesses each request component, it is necessary to determine whether each infringing component supports header bidding. If it does, it is necessary to further check whether there is an SDK access method. The SDK access method is used to request third-party advertising channel providers, that is, the third-party channel provider only provides the SDK access method. Some logic codes of the SDK access method are placed on the client. Therefore, when the client requests the advertising system, the advertising system sends the corresponding identifier RID (non-advertising data) to trigger the corresponding logic. The client can then invoke the SDK of the third-party advertising channel provider to request advertising data, and bring up advertising materials, prices and other data during the second request.

[0111] In the SDK access method, the client acts as a bridge between the third-party advertising channel and the advertising system. At this time, the client not only renders the advertisement to display to the user, but also acts as a data channel to call the SDK to obtain the advertisement data and give it to the advertising system. Therefore, when the client (user device) plays two roles, it can be found that after obtaining the advertising material, it is necessary to make a second request to the advertising system and carry the corresponding advertising material and price to the advertising system. The price carried in the second request is given by the third-party advertising channel.

[0112] See also Figure 4 , Figure 4 This is a schematic diagram of advertising traffic allocation corresponding to the hybrid allocation model in this embodiment. In this embodiment, when the traffic allocation model is a hybrid allocation model, advertising requests are allocated according to the traffic allocation links corresponding to each request component; if the type of the current request component is the first request component and it responds to the advertising request and fills the corresponding resource position with advertising resources, the filled advertising resources are returned to the client for advertising delivery, and the remaining traffic of the advertising request flows into the request component of the next priority advertising channel to continue advertising filling; if the type of the current request component is the second request component and it has the highest bid among the same priority levels, and the current request component responds to the advertising request and fills the corresponding resource position with advertising resources, the filled advertising resources are returned to the client for advertising delivery, and the remaining traffic of the advertising request flows into the request component of the next priority advertising channel to continue advertising filling.

[0113] like Figure 4 As shown, the hybrid allocation model includes the request components of each advertising channel (request components 1-5) and the traffic allocation links corresponding to each advertising channel (request component 1->request component 2-4->request component 5). That is, the hybrid allocation model is mainly based on waterfall flow and performs local real-time price comparison. Figure 4 There are three levels in total, namely: Level 1 (request component 1), Level 2 (request components 2, 3, 4) and Level 3 (request component 5). The essence of judging the level is to see whether there is a possibility of reducing the scale of traffic flowing into the level. Each level must have two arrow flows, one is the link where the filled ad data flows directly to the ad return link, and the other is the direction where the unfilled traffic flows to the next level. For example, the homepage carousel requires three ads to be rotated. The ad traffic initiated by the client first flows into Level 1 (request component 1). Assuming that only one ad is filled in request component 1, the remaining ad traffic flows into Level 2. At this time, any component in request components 2, 3, and 4 can be filled with a maximum of two ads. The bids of each advertising channel are asked and auctioned, and the highest bidder wins. If there is a failed auction, the remaining tail volume flows into Level 3 (request component 5). If there are other request components, the same applies. Figure 4 The reason why dotted lines are used to describe the traffic flow is that the traffic allocation strategy for each resource position, or even for different requests for the same resource position, may be different, and the traffic allocation model will be dynamically assembled. Figure 4 This is just a description of one of the traffic distribution strategies.

[0114] There are two types of general request components, one is the normal mode (only supports waterfall mode), and the other supports header bidding mode. Request components that support header bidding mode can be assembled using either waterfall assembly mode or hierarchical flattening assembly mode. The request components in normal mode only support waterfall mode assembly. The traffic allocation strategy is configured according to the specific business scenario. For example, for whether to access channel 1, it may be accessible in business scenario 1, but not in business scenario 2 (a typical case is that advertising channel 1 and the traffic party only signed a contract to access business scenario 1). Not only that, the entire traffic allocation model is dynamically adjusted according to the business scenario, who flows in first, who flows in next, or which channels compare prices in real time, etc.

[0115] Figure 4 It is a full-link hybrid distribution solution, that is, based on the main line waterfall flow distribution model, it carries a local real-time price comparison distribution model, and of course supports real-time price comparison for all request components. Therefore, the traffic distribution strategy can be flexibly configured, and different configurations result in different distribution strategies. Figure 4 This is just one of many traffic distribution strategies. Figure 4 The hybrid allocation model in is compatible with the SDK access method. Request components 2, 3, and 4 are at the same level, so the same traffic scale flows in. There is no upper and lower level relationship, that is, there is no priority.

[0116] The above describes the advertising flow distribution method in the embodiment of the present invention. The following describes the advertising flow distribution device in the embodiment of the present invention. Figure 5 In one embodiment of the present invention, an advertisement flow distribution device includes:

[0117] The traffic receiving module 501 is used to receive an advertisement request initiated by a client, wherein the advertisement request includes resource location information;

[0118] A model building module 502 is used to build a traffic allocation model corresponding to the current business scenario based on the resource position information, wherein the traffic allocation model includes request components of several advertising channels and corresponding traffic allocation links;

[0119] The traffic distribution module 503 is used to flow the advertisement request into the request component of the corresponding advertisement channel according to the traffic distribution link for advertisement filling based on the traffic distribution model.

[0120] Optionally, in one embodiment, the model building module 502 includes:

[0121] A determination unit 5021, configured to determine a service scenario corresponding to the current advertisement traffic allocation based on the resource position information;

[0122] A configuration unit 5022 is used to configure a traffic distribution strategy corresponding to the business scenario;

[0123] The construction unit 5023 is used to construct a traffic distribution model corresponding to the current business scenario based on the traffic distribution strategy.

[0124] Optionally, in one embodiment, the construction unit 5023 is specifically used for:

[0125] Based on the traffic distribution strategy, determining an assembly mode of request components of each advertising channel;

[0126] Based on the assembly mode, the request components of each advertising channel are dynamically assembled to obtain the traffic allocation link corresponding to each advertising channel, so as to form a traffic allocation model corresponding to the current business scenario.

[0127] Optionally, in one embodiment, the assembly mode includes a waterfall assembly mode and a level-flattening assembly mode;

[0128] The types of the request components include a first request component that only supports the waterfall assembly mode and a second request component that supports both the waterfall assembly mode and the level-flattening assembly mode.

[0129] Optionally, in one embodiment, the traffic allocation model includes a waterfall allocation model, a real-time price comparison allocation model, and a hybrid allocation model;

[0130] Among them, each request component in the waterfall flow allocation model adopts the waterfall flow mode for traffic allocation; each request component in the real-time price comparison allocation model adopts the header bidding mode for traffic allocation; each request component in the hybrid allocation model mainly uses the waterfall flow mode for traffic allocation, and partially adopts the header bidding mode for traffic allocation.

[0131] Optionally, in one embodiment, the traffic distribution module 503 is specifically used to:

[0132] When the traffic allocation model is the waterfall allocation model, the advertisement request is allocated according to the traffic allocation link corresponding to each request component. If the current request component responds to the advertisement request and fills the corresponding resource position with advertisement resources, the filled advertisement resources are returned to the client for advertisement delivery, and the remaining traffic of the advertisement request flows into the request component of the next priority advertisement channel to continue advertisement filling, wherein different advertisement channels correspond to different traffic allocation priorities;

[0133] When the traffic allocation model is the real-time price comparison allocation model, the advertising request is allocated according to the traffic allocation link corresponding to each request component. If there is currently a request component with the highest bid at the same priority level that responds to the advertising request and fills the corresponding resource position with advertising resources, the filled advertising resources are returned to the client for advertising delivery, wherein all advertising channels correspond to the same traffic allocation priority.

[0134] Optionally, in one embodiment, the traffic distribution module 503 is further configured to:

[0135] When the traffic distribution model is the hybrid distribution model, the advertisement request is distributed according to the traffic distribution link corresponding to each request component;

[0136] If the type of the current request component is the first request component and it responds to the advertisement request and fills the corresponding resource position with advertisement resources, the filled advertisement resources are returned to the client for advertisement delivery, and the remaining traffic of the advertisement request is flowed into the request component of the next priority advertisement channel to continue advertisement filling;

[0137] If the type of the current request component is the second request component and the bid is the highest among those with the same priority, and the current request component responds to the advertising request and fills the corresponding resource positions with advertising resources, the filled advertising resources are returned to the client for advertising delivery, and the remaining traffic of the advertising request is flowed into the request component of the next priority advertising channel to continue advertising filling.

[0138] Since the embodiments of the device part correspond to the embodiments of the above-mentioned method, please refer to the above-mentioned method embodiments for the introduction of the advertising traffic distribution device provided by the present invention. The present invention will not be repeated here, and it has the same beneficial effects as the above-mentioned advertising traffic distribution method.

[0139] above Figure 5 The advertisement traffic distribution device in the embodiment of the present invention is described in detail from the perspective of modular functional entities, and the computer device in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0140] Figure 66 is a schematic diagram of the structure of a computer device provided by an embodiment of the present invention. The computer device 600 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 610 (for example, one or more processors) and a memory 620, and one or more storage media 630 (for example, one or more mass storage devices) storing application programs 633 or data 632. Among them, the memory 620 and the storage medium 630 can be short-term storage or permanent storage. The program stored in the storage medium 630 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the computer device 600. Furthermore, the processor 610 can be configured to communicate with the storage medium 630 to execute a series of instruction operations in the storage medium 630 on the computer device 600.

[0141] The computer device 600 may also include one or more power supplies 640, one or more wired or wireless network interfaces 650, one or more input and output interfaces 660, and / or one or more operating systems 631, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. It will be appreciated by those skilled in the art that Figure 6 The illustrated computer device structure does not constitute a limitation on the computer device, and may include more or fewer components than illustrated, or combine certain components, or arrange the components differently.

[0142] The present invention also provides a computer device, which includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor executes the steps of the advertising traffic distribution method in the above-mentioned embodiments.

[0143] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the advertising traffic distribution method.

[0144] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0145] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0146] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for allocating advertising traffic, characterized in that: include: Receiving an advertisement request initiated by a client, wherein the advertisement request includes resource location information; Based on the resource location information, a traffic allocation model corresponding to the current business scenario is constructed, wherein the traffic allocation model includes request components of several advertising channels and corresponding traffic allocation links; Based on the traffic distribution model, the advertisement request is flowed into the request component of the corresponding advertisement channel according to the traffic distribution link for advertisement filling.

2. The advertising traffic distribution method according to claim 1, characterized in that: The constructing of a traffic distribution model corresponding to the current business scenario based on the resource location information includes: Based on the resource position information, determine the business scenario corresponding to the current advertising traffic allocation; Configure the traffic distribution strategy corresponding to the business scenario; Based on the traffic allocation strategy, a traffic allocation model corresponding to the current business scenario is constructed.

3. The advertising traffic distribution method according to claim 2, characterized in that: The constructing of a traffic distribution model corresponding to the current business scenario based on the traffic distribution strategy includes: Based on the traffic distribution strategy, determining an assembly mode of request components of each advertising channel; Based on the assembly mode, the request components of each advertising channel are dynamically assembled to obtain the traffic allocation link corresponding to each advertising channel, so as to form a traffic allocation model corresponding to the current business scenario.

4. The advertising traffic distribution method according to claim 3, characterized in that: The assembly modes include a waterfall assembly mode and a hierarchical flattening assembly mode; The types of the request components include a first request component that only supports the waterfall assembly mode and a second request component that supports both the waterfall assembly mode and the level-flattening assembly mode.

5. The advertising traffic distribution method according to any one of claims 1 to 4, characterized in that: The traffic allocation model includes a waterfall allocation model, a real-time price comparison allocation model and a mixed allocation model; Among them, each request component in the waterfall flow allocation model adopts the waterfall flow mode for traffic allocation; each request component in the real-time price comparison allocation model adopts the header bidding mode for traffic allocation; each request component in the hybrid allocation model mainly uses the waterfall flow mode for traffic allocation, and partially adopts the header bidding mode for traffic allocation.

6. The method for distributing advertising traffic according to claim 5, characterized in that: The step of flowing the advertisement request into the request component of the corresponding advertisement channel according to the traffic distribution link for advertisement filling based on the traffic distribution model includes: When the traffic allocation model is the waterfall allocation model, the advertisement request is allocated according to the traffic allocation link corresponding to each request component. If the current request component responds to the advertisement request and fills the corresponding resource position with advertisement resources, the filled advertisement resources are returned to the client for advertisement delivery, and the remaining traffic of the advertisement request flows into the request component of the next priority advertisement channel to continue advertisement filling, wherein different advertisement channels correspond to different traffic allocation priorities; When the traffic allocation model is the real-time price comparison allocation model, the advertising request is allocated according to the traffic allocation link corresponding to each request component. If there is currently a request component with the highest bid at the same priority level that responds to the advertising request and fills the corresponding resource position with advertising resources, the filled advertising resources are returned to the client for advertising delivery, wherein all advertising channels correspond to the same traffic allocation priority.

7. The method for distributing advertising traffic according to claim 5, characterized in that: The step of flowing the advertisement request into the request component of the corresponding advertisement channel according to the traffic distribution link for advertisement filling based on the traffic distribution model further comprises: When the traffic distribution model is the hybrid distribution model, the advertisement request is distributed according to the traffic distribution link corresponding to each request component; If the type of the current request component is the first request component and it responds to the advertisement request and fills the corresponding resource position with advertisement resources, the filled advertisement resources are returned to the client for advertisement delivery, and the remaining traffic of the advertisement request is flowed into the request component of the next priority advertisement channel to continue advertisement filling; If the type of the current request component is the second request component and the bid is the highest among those with the same priority, and the current request component responds to the advertising request and fills the corresponding resource positions with advertising resources, the filled advertising resources are returned to the client for advertising delivery, and the remaining traffic of the advertising request is flowed into the request component of the next priority advertising channel to continue advertising filling.

8. An advertising traffic distribution device, characterized in that: The advertising flow distribution device comprises: A traffic receiving module, configured to receive an advertisement request initiated by a client, wherein the advertisement request includes resource location information; A model building module, used to build a traffic allocation model corresponding to the current business scenario based on the resource position information, wherein the traffic allocation model includes request components of several advertising channels and corresponding traffic allocation links; The traffic distribution module is used to flow the advertisement request into the request component of the corresponding advertisement channel according to the traffic distribution link for advertisement filling based on the traffic distribution model.

9. A computer device, characterized in that: The computer device comprises: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory to enable the computer device to execute the advertising traffic distribution method according to any one of claims 1 to 7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the advertising traffic distribution method according to any one of claims 1 to 7 is implemented.

Images