CN105574741A - Flow value assessment method, device, application method and system - Google Patents

Abstract

The invention relates to the field of the Internet, and discloses a traffic value evaluation method, device, and application method and system thereof, so as to solve the technical problem that the traffic value cannot be accurately evaluated in the prior art. The method includes: providing an evaluation method for traffic value, including: obtaining multiple pieces of access information for predetermined network access objects within a historical time period; determining feature-related information of multiple feature types corresponding to each piece of access information; based on The feature-related information of each piece of access information constructs a feature vector corresponding to the access information, and then obtains multiple feature vectors; input the multiple feature vectors into the neural network model, and obtain the output value of the neural network model as the traffic value of the predetermined network access object, where , the neural network model is obtained by neural network language training based on multiple sampling samples. The technical effect of being able to accurately evaluate the traffic value of a predetermined network access object is achieved.

Description

A flow value evaluation method, device and application method and system thereof

technical field

The present invention relates to the technical field of the Internet, in particular to an evaluation method and device for traffic value and its application method and system.

Background technique

With the continuous development of science and technology, electronic technology has also developed rapidly. With the rise of the Internet, online advertising has become the main profit of major portals, search engines, social networks and applications of various terminal communication devices. Way. At present, the Internet advertising market is growing at an alarming rate, and the effectiveness of Internet advertising is becoming more and more important.

However, since the number of advertising slots is limited, usually the DSP (Demand-SidePlatform: Internet advertising) platform needs to determine the advertisement provided by which DSP platform to place on a certain advertising slot through bidding. , it is necessary to accurately estimate the traffic value of the website, so as to provide a powerful and reliable reference basis and decision-making information for the advertising party to bid. Accurate estimates cannot further provide a good data reference basis for improving the effectiveness of advertising.

Contents of the invention

In view of the above problems, the present invention is proposed to provide a traffic value evaluation method, device, and application method and system that overcome the above problems or at least partially solve the above problems.

In the first aspect, an embodiment of the present invention provides a traffic value evaluation method, including:

Obtain multiple pieces of access information for predetermined network access objects within a historical time period;

Determine feature-related information of multiple feature types corresponding to each piece of access information;

Constructing a feature vector corresponding to the access information based on the feature-related information of each piece of access information, and then obtaining multiple feature vectors;

Inputting the plurality of feature vectors into a neural network model to obtain an output value of the neural network model as the traffic value of the predetermined network access object, wherein the neural network model is obtained by performing neural network language training based on a plurality of sampling samples , each of the plurality of sampling samples includes multiple pieces of access information of a network access object and a traffic value of the corresponding network access object.

Optionally, the feature type includes at least any of the following:

access user;

access context;

place of visit;

interview time.

Optionally, the feature-related information of the accessing user includes any of the following:

User attribute information;

User's historical interest information;

User's real-time interest information.

Optionally, the feature-related information of the access context includes any of the following:

Terminal related information;

Internet-related information;

Ads display relevant information.

Optionally, the terminal-related information includes: at least one of the terminal model, the terminal operating system type, the terminal operating system version, and application information installed on the terminal.

Optionally, the predetermined network access object is a predetermined website or a predetermined application program.

In the second aspect, an embodiment of the present invention provides an advertisement placement method, including:

determining multiple target network access objects and multiple pieces of access information for each target network access object;

The method for evaluating traffic value introduced in any embodiment of the present invention is to determine the traffic value of each target network access object in the plurality of target network access objects;

Screen out the target network access objects whose traffic value ranks in the top M positions from the plurality of target network access objects, where M is a positive integer;

Advertisement is delivered to the target network access objects ranked in the top M positions; or, the target network access objects ranked in the top M positions are provided to an advertisement delivery party.

Optionally, the filtering out the target network access objects whose traffic value ranks in the top M positions from the plurality of target network access objects includes:

sorting the plurality of target network access objects according to the traffic value from high to low;

Obtain the target network access objects whose ranking is in the top M position.

Optionally, the filtering out the target network access objects whose traffic value ranks in the top M positions from the plurality of target network access objects includes:

judging whether the traffic value of each target network access object is greater than a preset threshold;

When the traffic value of the corresponding target network access object is greater than the preset threshold, the corresponding network access object is determined as one of the top M target network access objects in the ranking.

In a third aspect, an embodiment of the present invention provides an advertisement placement method, including:

Obtaining at least one target network access object and multiple pieces of access information of each target network access object;

The traffic value evaluation method introduced in any embodiment of the present invention determines the traffic value of each target network access object;

Based on the traffic value of the target network access object, determine the expenditure when placing advertisements on the target network access object; or provide the traffic value of the target network access object to the advertisement delivery party for the advertisement delivery party Based on the traffic value of the target network access object, determine the expenditure when advertising is placed on the target network access object.

Optionally, after the determination of the expenses for placing advertisements on the target network access objects, the method further includes:

Determining the advertising click-through rate CTR of the target network access object;

Determine the pay-per-click (CPC) of the advertisement of the target network access object, and the CPC is the expenditure;

By subtracting the product of the CTR and the CPC from the traffic value of the target network access object, the actual obtained value of advertisement placement on the corresponding target network access object is obtained.

In a fourth aspect, an embodiment of the present invention provides an evaluation device for traffic value, including:

An acquisition module, configured to acquire multiple pieces of access information for predetermined network access objects within a historical time period;

A first determination module, configured to determine feature-related information of multiple feature types corresponding to each piece of access information;

Constructing a model for constructing a feature vector corresponding to the access information based on the feature-related information of each piece of access information, and then obtaining multiple feature vectors;

A first obtaining module, configured to input the plurality of feature vectors into a neural network model, and obtain an output value of the neural network model as the traffic value of the predetermined network access object, wherein the neural network model is based on a plurality of samples The samples are obtained through neural network language training, and each of the multiple sampling samples includes multiple pieces of access information of a network access object and the traffic value of the corresponding network access object.

Optionally, the feature type includes at least any of the following:

access user;

access context;

place of visit;

interview time.

Optionally, the feature-related information of the accessing user includes any of the following:

User attribute information;

User's historical interest information;

User's real-time interest information.

Optionally, the feature-related information of the access context includes any of the following:

Terminal related information;

Internet-related information;

Ads display relevant information.

Optionally, the terminal-related information includes: at least one of the terminal model, the terminal operating system type, the terminal operating system version, and application information installed on the terminal.

Optionally, the predetermined network access object is a predetermined website or a predetermined application program.

In the fifth aspect, an embodiment of the present invention provides an advertisement delivery system, including:

The second determination module is configured to determine multiple target network access objects and multiple pieces of access information for each target network access object;

The traffic value evaluation device introduced in any embodiment of the present invention is used to determine the traffic value of each target network access object in the plurality of target network access objects;

A screening module, configured to filter out the target network access objects whose traffic value ranks in the top M positions from the plurality of target network access objects, where M is a positive integer;

A module is provided for delivering advertisements to the target network access objects ranked in the top M positions; or, providing the target network access objects ranked in the top M positions to an advertisement delivery party.

Optionally, the screening module includes:

a sorting unit, configured to sort the multiple target network access objects according to the traffic value from high to low;

The acquisition unit is configured to acquire the target network access objects whose ranking is in the top M positions.

Optionally, the screening module includes:

A judging unit, configured to judge whether the traffic value of each target network access object is greater than a preset threshold;

A determining unit, configured to determine the corresponding network access object as one of the top M target network access objects in the ranking when the traffic value of the corresponding target network access object is greater than the preset threshold.

In a sixth aspect, an embodiment of the present invention provides an advertisement delivery system, including:

The second obtaining module is used to obtain at least one target network access object and multiple pieces of access information of each target access object;

The flow value evaluation device introduced in any embodiment of the present invention is used to determine the flow value of each target network access object;

The third determining module is configured to determine the expenditure when the target network access object places advertisements based on the traffic value of the target network access object; or provide the advertisement delivery party with the traffic value of the target network access object, It is used for the advertisement delivery party to determine the expenditure when placing advertisements on the target network access object based on the traffic value of the target network access object.

Optionally, the system also includes:

The fourth determining module is used to determine the advertising click-through rate CTR of the target network access object;

The fifth determining module is used to determine the pay-per-click (CPC) of the advertisement of the target network access object, and the CPC is the expenditure;

The third obtaining module is used for subtracting the product of the CTR and the CPC from the traffic value of the target network access object to obtain the actual acquisition value of advertising delivery on the corresponding target network access object.

The technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

Since in the embodiment of the present invention, neural network language training can be performed based on a plurality of sampling samples to obtain a neural network model, each sampling sample in the plurality of sampling samples includes a plurality of pieces of access information of a network access object and a corresponding network access object The traffic value; and then for the predetermined network access object, multiple pieces of access information for the predetermined network access object can be obtained in the historical time period, and the feature-related information of various feature types corresponding to each piece of access information can be determined, and then based on each The feature-related information of the access information constructs the feature vector corresponding to the access information, and then obtains multiple feature vectors. Finally, input the pre-trained neural network model into the multiple feature vectors to obtain the traffic value of the predetermined network access object, which can achieve The technical effect of accurately assessing the traffic value of intended network access objects.

The above description is only an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention, it can be implemented according to the contents of the description, and in order to make the above and other purposes, features and advantages of the present invention more obvious and understandable , the specific embodiments of the present invention are enumerated below.

Description of drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiment. The drawings are only for the purpose of illustrating a preferred embodiment and are not to be considered as limiting the invention. Also throughout the drawings, the same reference numerals are used to designate the same components. In the attached picture:

FIG. 1 shows a flowchart of a flow value evaluation method according to an embodiment of the present invention;

Fig. 2 shows a flow chart of an advertisement delivery method according to an embodiment of the present invention;

Fig. 3 shows a flowchart of an advertisement placement method according to another embodiment of the present invention;

FIG. 4 shows a structural diagram of an evaluation device for traffic value according to an embodiment of the present invention;

Fig. 5 shows a structural diagram of an advertisement delivery system according to an embodiment of the present invention;

Fig. 6 shows a structural diagram of an advertisement delivery system according to another embodiment of the present invention.

detailed description

The general idea of the scheme in the embodiment of the present invention is as follows:

An evaluation method for traffic value is provided, including: obtaining multiple pieces of access information for predetermined network access objects within a historical time period; determining feature-related information of various feature types corresponding to each piece of access information; based on each piece of access information The feature-related information of the corresponding access information constructs a feature vector corresponding to the access information, and then obtains multiple feature vectors; enters multiple feature vectors into the neural network model, and obtains the output value of the neural network model as the traffic value of the predetermined network access object, wherein the neural network model The neural network language training is performed based on multiple sampling samples, each of the multiple sampling samples includes multiple pieces of access information of a network access object and the traffic value of the corresponding network access object. The technical effect of being able to accurately evaluate the traffic value of a predetermined network access object is achieved.

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided for more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

An embodiment of the present invention provides a traffic value evaluation method, please refer to Figure 1, including:

Step S101: Obtain multiple pieces of access information for predetermined network access objects within a historical time period;

Step S102: Determine feature-related information of multiple feature types corresponding to each piece of access information;

Step S103: Construct a feature vector corresponding to the access information based on the feature-related information of each piece of access information, and then obtain multiple feature vectors;

Step S104: Input multiple feature vectors into the neural network model, and obtain the output value of the neural network model as the traffic value of the predetermined network access object, wherein the neural network model is obtained by performing neural network training based on multiple sampling samples, and among the multiple sampling samples Each sampling sample of includes multiple pieces of access information of a network access object and the traffic value of the corresponding network access object.

In step S101, different historical time periods can be set according to actual needs, for example: 1 month, 3 months, etc., which are not limited in this embodiment of the present invention.

The reserved network access objects are, for example: reserved websites, reserved APPs, and the like.

The access information for a predetermined network access object includes, for example, browsing information, click information, information about downloading an APP provided by it, etc., which are not limited in this embodiment of the present invention.

In step S102, the characteristic types include but not limited to: visiting user, visiting context, visiting location, visiting time and so on. The access location is, for example, the location of the terminal when accessing a specific network access object; the access time is, for example, the time of accessing a specific network access object.

Among them, the information related to the characteristics of the accessing user includes but is not limited to:

User attribute information, such as gender, age, education level, income, occupation information, etc.;

The user's historical interest information, such as the advertisement information clicked or viewed by the user, the web page information browsed by the user, and the purchase information of the user's history;

The user's real-time interest information, for example, the information of the web page currently browsed by the user.

Among them, the feature-related information of the access context includes but is not limited to:

Terminal-related information, such as terminal model, operating system type, operating system version, and application information installed on the terminal.

Network-related information, such as network operator information and network type of the terminal connected to the network;

Ad display-related information, such as ad size information, ad placement, and so on.

In step S103, for each piece of access information, a certain value may be assigned to each feature-related information, and then a feature vector of the access information may be constructed. Wherein, each feature-related information constitutes one-dimensional data in the feature vector, and then how many feature-related information of the access information is collected, and how many-dimensional feature vectors corresponding to the access information can be obtained.

In step S104, the multiple feature vectors corresponding to the obtained pieces of access information are directly input into the neural network model, and the output result of the obtained neural network model is the traffic value of the predetermined network access object.

The neural network model is, for example, a BP (BackPropagation: reverse propagation) neural network model. BP neural network is a multilayer feed-forward network trained by error backpropagation algorithm, and it is one of the most widely used neural network models at present. The BP network can learn and store a large number of input-output pattern mapping relationships without revealing the mathematical equations describing the mapping relationship in advance. Its learning rule is to use the steepest descent method to continuously adjust the weights and thresholds of the network through backpropagation to minimize the sum of squared errors of the network. The topological structure of BP neural network model includes input layer (input), hidden layer (hidelayer) and output layer (outputlayer).

The present invention also discloses a method for placing advertisements, which is a further utilization of the aforementioned flow value evaluation method, please refer to Figure 2, which includes the following steps:

Step S201: Determine multiple target network access objects and multiple pieces of access information for each target network access object;

Step S202: Based on the traffic value evaluation method introduced in any embodiment of the present invention, determine the traffic value of each target network access object among the multiple target network access objects;

Step S203: Screen out the target network access objects whose traffic value ranks in the top M positions from the multiple target network access objects, where M is a positive integer;

Step S204: delivering advertisements to the target network access objects ranked in the top M positions; or, providing the target network access objects ranked in the top M positions to the advertisement delivery party.

In step S203, multiple target network access objects can be screened in various ways, two of which are listed below for introduction, of course, in the specific implementation process, it is not limited to the following two situations.

The first method is to filter out the target network access objects whose traffic value ranks in the top M positions from multiple target network access objects, including: sorting the multiple target network access objects according to the traffic value from high to low; The M-bit target network access object.

For example, the value of M is fixed first, and the value of M can be set directly, such as: 3, 5, etc., or the number of multiple target network access objects is determined first, and then the number of the number is taken. A predetermined ratio (for example: 20%, 30%, etc.) is used as the value of M, and the method for determining the value of M is not limited in this embodiment of the present invention. Then, by sorting the multiple target network access objects according to the traffic value from high to low, the traffic value ranked in the top M positions can be obtained.

The second method is to filter out the target network access objects whose traffic value ranks in the top M positions from multiple target network access objects, including: judging whether the traffic value of each target network access object is greater than the preset threshold; When the traffic value of the object is greater than the preset threshold, the corresponding target network access object is determined as one of the top M target network access objects.

For example, the value of M is not set in advance, but the traffic value of each target website visitor is directly compared with the preset traffic value, if the traffic value of the corresponding target network visitor is greater than the preset traffic value, it is used as one of the target network access objects ranked in the top M positions, so as to finally determine M variable target network access objects.

In step S203, since the traffic value of the target network access objects ranked in the top M positions is relatively high, the click-through rate of the advertisements will be relatively high after the advertisements are delivered to them, thereby improving the efficiency of advertisement delivery. Wherein, the advertisement can be delivered directly to the predetermined network access objects, or the target network access objects ranked in the top M positions can be provided to other advertisement delivery parties, and then other advertisement delivery parties can place advertisements on these target network access objects.

The advertising provider is, for example, a DSP (Demand-Side Platform: advertising demand side) platform. The DSP platform is an online advertising platform that serves advertisers and helps advertisers place advertisements on the Internet or mobile Internet. DSP can make advertisements It is easier and more convenient for the owner to follow the unified bidding and feedback method, and purchase high-quality advertising inventory in real time at a reasonable price for online advertisements located on multiple advertising exchange platforms.

Among them, after the DSP platform determines the top M target network visitors, it believes that the return of advertising on these target network visitors is relatively high, so it can place advertisements on these target network visitors through bidding.

The present invention also discloses another advertising delivery method, which is a further utilization of the aforementioned flow value evaluation method, please refer to Figure 3, which includes the following steps:

Step S301: Obtain at least one target network access object and multiple pieces of access information of each target access object;

Step S302: Determine the traffic value of each target network access object based on the traffic value evaluation method introduced in any embodiment of the present invention;

Step S303: Based on the traffic value of the target network access object, determine the expenditure when placing advertisements on the target network access object; or provide the advertisement delivery party with the traffic value of the target network access object for the advertisement delivery party to use based on the target network access object The value of traffic to determine how much to spend when serving ads to target web visitors.

In step S303, under normal circumstances, every time the advertisement of the target network visitor is clicked, it is necessary to pay a CPC (CostPerClick: Advertisement Pay Per Click) to the provider of the target network visitor, and at the same time, the cost of the target network visitor can be roughly estimated. CTR (Click-Through-Rate: ad click-through rate), if CPC*CTR is less than the traffic value, then the advertising party can get benefits, therefore, the advertising party can divide the traffic value by/estimated CTR to determine the The amount spent when an ad is served to targeted web visitors. For example: the spend fee is slightly less than the traffic value divided by/estimated CTR.

Based on this solution, it can be ensured that the advertising provider can accurately determine the upper limit of expenditure when placing advertisements on target network access objects, thereby providing a good data reference basis for improving the efficiency of advertising delivery.

As an optional embodiment, after the determination of the expenditures for placing advertisements on the target network access object, the method further includes: determining the advertisement click-to-reach rate CTR of the target network access object; determining the Pay-per-click CPC of the advertisement of the target network access object, and the CPC is the expenditure cost; subtract the product of the CTR and the CPC from the traffic value of the target network access object to obtain the corresponding The actual earned value of an ad served to a targeted web audience.

Wherein, the CTR of the target network access object can be determined by analyzing the historical advertisements of the target network access object, based on the number of traffic and clicks of the historical advertisement. Then, get the advertising cost of the advertiser when placing the advertisement as the CPC, based on CPC*CTR, you can determine the advertising cost of the advertiser, and subtract the advertising cost from the traffic value, you can determine the amount of traffic in the corresponding target network. The actual earned value of the object's ad serving.

In addition, the ROL (ReturnOnInvestment: return on investment) of the advertising party placing the advertisement on the target network visitor can be obtained by dividing the actual obtained value by the advertisement serving cost.

Based on the traffic value evaluation method provided by the embodiment of the present invention, the embodiment of the present invention also provides a traffic value evaluation device, please refer to Figure 4, including:

An acquisition module 40, configured to acquire multiple pieces of access information for predetermined network access objects within a historical time period;

The first determination module 41 is configured to determine feature-related information of multiple feature types corresponding to each piece of access information;

Constructing a model 42, configured to construct a feature vector corresponding to the access information based on the feature-related information of each piece of access information, and then obtain multiple feature vectors;

The first obtaining module 43 is used to input multiple feature vectors into the neural network model, and obtain the output value of the neural network model as the traffic value of the predetermined network access object, wherein the neural network model is obtained by neural network language training based on a plurality of sampling samples , each of the plurality of sampling samples includes multiple pieces of access information of a network access object and a traffic value of the corresponding network access object.

Optionally, the feature type includes at least any one of the following: access user; access context; access location; access time.

Optionally, the feature-related information of the accessed user includes any of the following: user attribute information; user historical interest information; user real-time interest information.

Optionally, the feature-related information of the access context includes any of the following: terminal-related information; network-related information; advertisement display-related information.

Optionally, the terminal-related information includes: at least one of terminal model, terminal operating system type, terminal operating system version, and application program information installed on the terminal.

Optionally, the predetermined network access object is a predetermined website or a predetermined application program.

Due to the evaluation device of the above-mentioned traffic value, the device adopted for implementing the evaluation method of the traffic value introduced above, based on the evaluation method of the traffic value introduced in the embodiment of the present invention, those skilled in the art can understand the evaluation of the above-mentioned traffic value The specific structure and deformation of the device will not be repeated here, and all devices used to implement the aforementioned traffic value evaluation method belong to the protection scope of the embodiments of the present invention.

Based on the advertisement delivery method provided by the embodiment of the present invention, the embodiment of the present invention also provides an advertisement delivery system, please refer to Figure 5, including:

The second determination module 50 is configured to determine a plurality of target network access objects and multiple pieces of access information of each target network access object;

The traffic value evaluation device 51 introduced in any embodiment of the present invention is used to determine the traffic value of each target network access object among the multiple target network access objects;

The screening module 52 is used to filter out the target network access objects whose traffic value ranking is in the top M positions from a plurality of target network access objects, where M is a positive integer;

The providing module 53 is configured to deliver advertisements to the target network access objects ranked in the top M positions; or provide the target network access objects ranked in the top M positions to the advertisement delivery party.

Optionally, the screening module 52 includes: a sorting unit, configured to sort the multiple target network access objects according to the traffic value from high to low; an obtaining unit, configured to obtain the target network access objects ranked in the top M ranks.

Optionally, the screening module 52 includes: a judging unit, configured to judge whether the traffic value of each target network access object is greater than a preset threshold; a determination unit, configured to determine whether the traffic value of the corresponding target network access object is greater than a preset threshold , determining the corresponding network access object as one of the target network access objects ranked in the top M positions.

Due to the above-mentioned advertisement delivery system, the system used to implement the first advertisement delivery method introduced above, and based on the first advertisement delivery method introduced in the embodiment of the present invention, those skilled in the art can understand the above-mentioned advertisement delivery system. The specific structure and deformation will not be repeated here, and any system used to implement the aforementioned first advertising delivery method falls within the scope of protection intended by the embodiments of the present invention.

Based on the advertisement delivery method provided by the embodiment of the present invention, the embodiment of the present invention also provides an advertisement delivery system, please refer to Figure 6, including:

The second obtaining module 60 is configured to obtain at least one target network access object and multiple pieces of access information of each target access object;

The traffic value evaluation device 61 introduced in any embodiment of the present invention is used to determine the traffic value of each target network access object;

The third determination module 62 is used to determine the expenditure when advertising is placed on the target network access object based on the traffic value of the target network access object; or provide the traffic value of the target network access object to the advertisement delivery party for the advertisement delivery party Based on the traffic value of the target network visitors, the expenditures for advertising on the target network visitors are determined.

Optionally, the system also includes: a fourth determination module, used to determine the advertisement click-through rate CTR of the target network access object; a fifth determination module, used to determine the advertisement pay-per-click CPC of the target network access object, and the CPC is Expenses; the third acquisition module is used to subtract the product of CTR and CPC from the traffic value of the target network access object to obtain the actual acquisition value of advertising on the corresponding target network access object.

Due to the above-mentioned advertisement delivery system, the system used to implement the second advertisement delivery method introduced above, based on the second advertisement delivery method introduced in the embodiment of the present invention, those skilled in the art can understand the above-mentioned advertisement delivery system. The specific structure and deformation will not be repeated here, and any system used to implement the aforementioned second advertisement placement method falls within the scope of protection intended by the embodiments of the present invention.

The technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

Since in the embodiment of the present invention, neural network language training can be performed based on a plurality of sampling samples to obtain a neural network model, each sampling sample in the plurality of sampling samples includes a plurality of pieces of access information of a network access object and a corresponding network access object The traffic value; and then for the predetermined network access object, multiple pieces of access information for the predetermined network access object can be obtained in the historical time period, and the feature-related information of various feature types corresponding to each piece of access information can be determined, and then based on each The feature-related information of the access information constructs the feature vector corresponding to the access information, and then obtains multiple feature vectors. Finally, input the pre-trained neural network model into the multiple feature vectors to obtain the traffic value of the predetermined network access object, which can achieve The technical effect of accurately assessing the traffic value of intended network access objects.

The algorithms and displays presented herein are not inherently related to any particular computer, virtual system, or other device. Various generic systems can also be used with the teachings based on this. The structure required to construct such a system is apparent from the above description. Furthermore, the present invention is not specific to any particular programming language. It should be understood that various programming languages can be used to implement the content of the present invention described herein, and the above description of specific languages is for disclosing the best mode of the present invention.

In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure the understanding of this description.

Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, in order to streamline this disclosure and to facilitate an understanding of one or more of the various inventive aspects, various features of the invention are sometimes grouped together in a single embodiment, figure, or its description. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the Detailed Description are hereby expressly incorporated into this Detailed Description, with each claim standing on its own as a separate embodiment of this invention.

Those skilled in the art can understand that the modules in the device in the embodiment can be adaptively changed and arranged in one or more devices different from the embodiment. Modules or units or components in the embodiments may be combined into one module or unit or component, and furthermore may be divided into a plurality of sub-modules or sub-units or sub-assemblies. All features disclosed in this specification (including accompanying claims, abstract and drawings) and any method or method so disclosed may be used in any combination, except that at least some of such features and/or processes or units are mutually exclusive. All processes or units of equipment are combined. Each feature disclosed in this specification (including accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

Furthermore, those skilled in the art will understand that although some embodiments herein include some features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention. And form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.

The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all functions of some or all components in the gateway, proxy server, and system according to the embodiments of the present invention. The present invention can also be implemented as an apparatus or an apparatus program (for example, a computer program and a computer program product) for performing a part or all of the methods described herein. Such a program for realizing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet site, or provided on a carrier signal, or provided in any other form.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means can be embodied by one and the same item of hardware. The use of the words first, second, and third, etc. does not indicate any order. These words can be interpreted as names.

The present invention discloses, A1, an evaluation method of flow value, comprising:

Obtain multiple pieces of access information for predetermined network access objects within a historical time period;

Determine feature-related information of multiple feature types corresponding to each piece of access information;

Constructing a feature vector corresponding to the access information based on the feature-related information of each piece of access information, and then obtaining multiple feature vectors;

Inputting the plurality of feature vectors into a neural network model to obtain an output value of the neural network model as the traffic value of the predetermined network access object, wherein the neural network model is obtained by performing neural network language training based on a plurality of sampling samples , each of the plurality of sampling samples includes multiple pieces of access information of a network access object and a traffic value of the corresponding network access object.

A2. The method as described in A1, wherein the feature type includes at least any one of the following:

access user;

access context;

place of visit;

interview time.

A3, the method as described in A2, is characterized in that, the feature-related information of the access user includes any of the following:

User attribute information;

User's historical interest information;

User's real-time interest information.

A4. The method as described in A2, wherein the feature-related information of the access context includes any of the following:

Terminal related information;

Internet-related information;

Ads display relevant information.

A5. The method as described in A4, wherein the terminal-related information includes: the model of the terminal, the type of operating system of the terminal, the version of the operating system of the terminal, and the application program installed on the terminal at least one of the information.

A6. The method according to any one of A1-A5, wherein the predetermined network access object is a predetermined website or a predetermined application program.

B7. A method for advertising, comprising:

determining multiple target network access objects and multiple pieces of access information for each target network access object;

Based on the evaluation method of traffic value described in any one of claims A1-A6, determine the traffic value of each target network access object in the plurality of target network access objects;

Screen out the target network access objects whose traffic value ranks in the top M positions from the plurality of target network access objects, where M is a positive integer;

Advertisement is delivered to the target network access objects ranked in the top M positions; or, the target network access objects ranked in the top M positions are provided to an advertisement delivery party.

B8. The method as described in B7, wherein the filtering out the target network access objects whose traffic value ranking is in the top M positions from the plurality of target network access objects includes:

sorting the plurality of target network access objects according to the traffic value from high to low;

Obtain the target network access objects whose ranking is in the top M positions.

B9. The method as described in B7, wherein the filtering out the target network access objects whose traffic value ranking is in the top M positions from the plurality of target network access objects includes:

judging whether the traffic value of each target network access object is greater than a preset threshold;

When the traffic value of the corresponding target network access object is greater than the preset threshold, the corresponding network access object is determined as one of the top M target network access objects in the ranking.

C10. An advertisement delivery method, comprising:

Obtaining at least one target network access object and multiple pieces of access information of each target network access object;

Based on any of the traffic value evaluation methods described in A1-A6, determine the traffic value of each target network access object;

Based on the traffic value of the target network access object, determine the expenditure when placing advertisements on the target network access object; or provide the traffic value of the target network access object to the advertisement delivery party for the advertisement delivery party Based on the traffic value of the target network access object, determine the expenditure when advertising is placed on the target network access object.

C11. The method as described in C10, characterized in that, after the determination of the expenditure expenses when placing advertisements on the target network access object, the method also includes:

Determining the advertising click-through rate CTR of the target network access object;

Determine the pay-per-click (CPC) of the advertisement of the target network access object, and the CPC is the expenditure;

By subtracting the product of the CTR and the CPC from the traffic value of the target network access object, the actual obtained value of advertisement placement on the corresponding target network access object is obtained.

D12. An evaluation device for flow value, comprising:

An acquisition module, configured to acquire multiple pieces of access information for predetermined network access objects within a historical time period;

A first determination module, configured to determine feature-related information of multiple feature types corresponding to each piece of access information;

Constructing a model for constructing a feature vector corresponding to the access information based on the feature-related information of each piece of access information, and then obtaining multiple feature vectors;

A first obtaining module, configured to input the plurality of feature vectors into a neural network model, and obtain an output value of the neural network model as the traffic value of the predetermined network access object, wherein the neural network model is based on a plurality of samples The samples are obtained through neural network language training, and each of the multiple sampling samples includes multiple pieces of access information of a network access object and the traffic value of the corresponding network access object.

D13. The device as described in D12, wherein the feature type includes at least any one of the following:

access user;

access context;

place of visit;

interview time.

D14. The device as described in D13, wherein the feature-related information of the accessing user includes any of the following:

User attribute information;

User's historical interest information;

User's real-time interest information.

D15. The device according to D13, wherein the feature-related information of the access context includes any of the following:

Terminal related information;

Internet-related information;

Ads display relevant information.

D16. The device as described in D15, wherein the terminal-related information includes: the model of the terminal, the type of operating system of the terminal, the version of the operating system of the terminal, and the application program installed on the terminal at least one of the information.

D17. The device according to any one of D12-D16, wherein the predetermined network access object is a predetermined website or a predetermined application program.

E18. An advertisement delivery system, characterized in that it includes:

The second determination module is configured to determine multiple target network access objects and multiple pieces of access information for each target network access object;

The traffic value evaluation device according to any one of claims D12-D17, configured to determine the traffic value of each target network access object in the plurality of target network access objects;

A screening module, configured to filter out the target network access objects whose traffic value ranks in the top M positions from the plurality of target network access objects, where M is a positive integer;

A module is provided for delivering advertisements to the target network access objects ranked in the top M positions; or, providing the target network access objects ranked in the top M positions to an advertisement delivery party.

E19. The system as described in E18, wherein the screening module includes:

a sorting unit, configured to sort the multiple target network access objects according to the traffic value from high to low;

The acquisition unit is configured to acquire the target network access objects whose ranking is in the top M positions.

E20, the system as described in E18, is characterized in that, described screening module comprises:

A judging unit, configured to judge whether the traffic value of each target network access object is greater than a preset threshold;

A determining unit, configured to determine the corresponding network access object as one of the top M target network access objects in the ranking when the traffic value of the corresponding target network access object is greater than the preset threshold.

F21. An advertisement delivery system, comprising:

The second obtaining module is used to obtain at least one target network access object and multiple pieces of access information of each target access object;

The traffic value evaluation device described in any one of claims F12-F17, used to determine the traffic value of each target network access object;

The third determining module is configured to determine the expenditure when the target network access object places advertisements based on the traffic value of the target network access object; or provide the advertisement delivery party with the traffic value of the target network access object, It is used for the advertisement delivery party to determine the expenditure when placing advertisements on the target network access object based on the traffic value of the target network access object.

F22, the system as described in F21, is characterized in that, described system also comprises:

The fourth determining module is used to determine the advertising click-through rate CTR of the target network access object;

The fifth determining module is used to determine the pay-per-click (CPC) of the advertisement of the target network access object, and the CPC is the expenditure;

The third obtaining module is used for subtracting the product of the CTR and the CPC from the traffic value of the target network access object to obtain the actual acquisition value of advertising delivery on the corresponding target network access object.

Claims (10)

1. A method for assessing traffic value, comprising: Obtain multiple pieces of access information for predetermined network access objects within a historical time period; Determining feature-related information of multiple feature types corresponding to each piece of access information; Constructing a feature vector corresponding to the access information based on the feature-related information of each piece of access information, and then obtaining multiple feature vectors; Inputting the plurality of feature vectors into a neural network model to obtain an output value of the neural network model as the traffic value of the predetermined network access object, wherein the neural network model is obtained by performing neural network language training based on a plurality of sampling samples , each of the plurality of sampling samples includes multiple pieces of access information of a network access object and a traffic value of the corresponding network access object. 2. The method of claim 1, wherein the feature type includes at least any of the following: access user; access context; place of visit; interview time. 3. The method according to claim 2, wherein the feature-related information of the accessing user includes any of the following: User attribute information; User's historical interest information; User's real-time interest information. 4. The method according to claim 2, wherein the feature-related information of the access context includes any of the following: Terminal related information; Internet-related information; Ads display relevant information. 5. The method according to claim 4, wherein the terminal-related information includes: the model of the terminal, the type of operating system of the terminal, the version of the operating system of the terminal, the At least one type of information in the application information. 6. An advertisement delivery method, characterized in that it comprises: determining multiple target network access objects and multiple pieces of access information for each target network access object; Based on the evaluation method of traffic value according to any one of claims 1-5, determining the traffic value of each target network access object in the plurality of target network access objects; Screen out the target network access objects whose traffic value ranks in the top M positions from the plurality of target network access objects, where M is a positive integer; Advertisement is delivered to the target network access objects ranked in the top M positions; or, the target network access objects ranked in the top M positions are provided to an advertisement delivery party. 7. An advertisement delivery method, characterized in that it comprises: Obtaining at least one target network access object and multiple pieces of access information of each target network access object; Based on the traffic value evaluation method described in any one of claims 1-5, determining the traffic value of each target network access object; Based on the traffic value of the target network access object, determine the expenditure when placing advertisements on the target network access object; or provide the traffic value of the target network access object to the advertisement delivery party for the advertisement delivery party Based on the traffic value of the target network access object, determine the expenditure when advertising is placed on the target network access object. 8. An evaluation device for flow value, characterized in that it comprises: An acquisition module, configured to acquire multiple pieces of access information for predetermined network access objects within a historical time period; A first determination module, configured to determine feature-related information of multiple feature types corresponding to each piece of access information; Constructing a model for constructing a feature vector corresponding to the access information based on the feature-related information of each piece of access information, and then obtaining multiple feature vectors; A first obtaining module, configured to input the plurality of feature vectors into a neural network model, and obtain an output value of the neural network model as the traffic value of the predetermined network access object, wherein the neural network model is based on a plurality of samples The samples are obtained through neural network language training, and each of the multiple sampling samples includes multiple pieces of access information of a network access object and the traffic value of the corresponding network access object. 9. An advertisement delivery system, characterized in that it comprises: The second determining module is configured to determine multiple target network access objects and multiple pieces of access information for each target network access object; The traffic value evaluation device according to claim 8, configured to determine the traffic value of each target network access object in the plurality of target network access objects; A screening module, configured to filter out the target network access objects whose traffic value ranks in the top M positions from the plurality of target network access objects, where M is a positive integer; A module is provided for delivering advertisements to the target network access objects ranked in the top M positions; or, providing the target network access objects ranked in the top M positions to an advertisement delivery party. 10. An advertisement delivery system, characterized in that it comprises: The second obtaining module is used to obtain at least one target network access object and multiple pieces of access information of each target access object; The evaluation device of traffic value according to claim 8, which is used to determine the traffic value of each target network access object; The third determining module is configured to determine the expenditure when the target network access object places advertisements based on the traffic value of the target network access object; or provide the advertisement delivery party with the traffic value of the target network access object, It is used for the advertisement delivery party to determine the expenditure when placing advertisements on the target network access object based on the traffic value of the target network access object.