In order to make the objects, technical solutions, and advantages of the present invention, the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that these descriptions are merely exemplary and are not to limit the scope of the invention. Further, in the following description, a description of the well-known structure and techniques is omitted to avoid unnecessary obscuring the concepts of the present invention.
 The terms used in the present invention are only for the purpose of describing particular embodiments, not intended to limit the invention. "One", "one", "one", "" "and" "" as used in the present invention and the appended claims are also intended to include most forms unless the context clearly represents other meanings. It should also be understood that the terms "and / or" as used herein refer to any or more of any or all possible combinations of one or more associated listing items.
 It should be understood that various information may be described in terms of the terms first, second, third, etc., but this information should not be limited to these terms. These terms are used only to distinguish the same type of information from each other. For example, the first information may be referred to as a second information without departing from the scope of the invention, and the second information can also be referred to as the first information. Depending on the context, if the words used herein "If" can be interpreted as "in ..." or "when ..." or "in response to determination".
 See figure 1 In one embodiment, the management method based on large data and artificial intelligence digital advertising can include:
 S1, receive the advertisement delivery request sent by the advertisement, and acquire the characteristics of the product type, product price, and advertising budget in the advertising delivery request, to generate product type characteristics, product price characteristics, and placement budget characteristics, then product type characteristics, product Price characteristics and destination budget features for dual linear transformations and feature recombination to generate first play features.
 Alternatively, advertising delivery requests include product types, product prices and advertising budgets, product types for ads for ads, specific categories and names for products marketing, such as target advertising marketing products are potato chips or shoes in zero foods. High heels; product prices are sold prices, for example, target advertising marketing products are zero foods in the potato chips, selling prices are 5 yuan a package; advertising budget is the price of advertisers to pay for the target advertisement For example, the advertisement is willing to pay 1 million yuan for the target advertisement.
 Advertising main terminals are intelligent devices with communication functions and data transmission functions for advertisers, including: smartphones, laptops, tablets, and desktop computers. Advertising is a legal person, other economic organization or individuals, other economic organizations, or individuals to design, production, or entrusting others to market for goods or services. The advertiser is the publisher of the advertising campaign. It is a merchant selling or promoting its own products and services online is the provider of alliance marketing advertisements. Any businesses that sell their products or services can be used as advertisers. The advertisers are responsible for providing market and commodity materials to advertising planning platforms, overseeing and accepting advertisements.
S2, using different scales of convolution, consolidation of product type characteristics, product price characteristics, and launch budget characteristics, and will enumerate the product type characteristics, product price characteristics, and planting budget feature input double-layer convolution network for obtaining Different dimensions of product type features, product price characteristics and launch budget features, and then perform feature splicing in the characteristic space of different dimensions, product price characteristics, and launch budget features to obtain second placement.
 S3, feature reconstruction of the first delivery characteristics to generate the first deposit reconstruction feature, and perform feature reconstruction of the second destination to generate the second deposit reconstruction feature, and then the first investment reconstruction feature and the second Linear aggregation is performed to obtain linear polymerization.
 In one embodiment, feature reconstruction generation of the first delivery feature includes:
 The first placement feature is mapped to the feature vector space to obtain the first placement feature vector, and utilize the preset linear transform coefficient and the linear bias coefficient to linearly transform the first destination feature vector to generate linear delivery feature vectors;
 The linear destination feature vector is turned to obtain a transposition feature vector, and the linear delivery feature vector and the transposition feature vector are brought into characteristic synthesis of linear delivery characteristics and transposition characteristics to obtain synthesis. The feature vector, then the synthetic feature vector is output to the weight feature vector to obtain a feature distribution vector;
 The value of each element in the feature distribution vector is normalized by feature compression functions to obtain a compressed feature vector; the feature compression function is a normalized index function;
 Feature decomposition of the first placement is characterized to obtain a number of first amplifier characteristics, and acquire correlation coefficients of each of the first casting loom characteristics based on the compressed feature vector;
 The first projection sub-feature of the correlation coefficient is greater than the preset threshold is used as the associated amphiplayer feature, and all first projection sub-features are fused to characterize the first casting reconstruction characteristic according to the correlation coefficient of each of the first projection loop features.
 In one embodiment, feature reconstruction generation of the second destination feature includes:
 Decompose the second placement as a number of second vapor features, traversing all second vaporizers, and will be traversed, the second projection loom feature is used as the target, and then acquire the target piton characteristics and other each The relationship between the second projection sub-characterization is characterized by a number of pastelation relationship sub-characteristics of the target.
 The polymerization of all the lamino characteristics of the target vapor characteristics is generated by the characteristic polymerization function to obtain a polymerization sub-characteristic of the target vaporized sub-feature; repeating the above steps to obtain polymerization subscriber characteristics of each second amplifier ;
 The polymerization sub-feature of all second projection sub-features is input to the second destinatory feature to obtain a second casting reconstruction feature.
 In one embodiment, linear polymerization of the first casting reconstructed feature and the second deposit reconstruction feature is obtained from advertising, including:
 Extract the correction characteristics of the first casting reconstruction feature and the second deposition reconstruction feature, and acquire the first via reconstructed feature and the second deposition reconstruction feature corresponding to the relationship characteristics to obtain the first correlation feature and the second Association characteristics;
 The first correlation feature and the second association feature are respectively mapped to the feature vector space of the same dimension to obtain the first associated feature vector and the second associated feature vector, and the first associated feature vector and the second associated feature vector are veneed. Get associated feature vectors;
 The associated feature vector is decomposed to obtain several associated sub-vectors, and each associated sub-vector is linearly transformed separately by different linear transform coefficients and linear transformation bias, respectively, respectively, respectively, respectively, to obtain several linear associated sub-vectors;
 All linear associated sub-vectors are polymerized to segment the associated feature vectors to obtain advertisements.
 S4, from the database acquire the advertising bit data, and extract the characteristics of the advertising bit data to obtain the advertisement timing characteristics of the advertising bits, then the ad caps timing characteristics are timed to obtain the timing characteristics of the advertisement and several advertising bits. feature.
 Advertising data includes type information, audience information, price information, and conversion rate information; the type information is the type of advertising, such as TV commercials, online advertising, and broadcast ads. The audience information is the audience of the advertising position, that is, the common features of the people who view the advertising are: youth, students and food, etc .; the price information is available for advertising sites at all cycles of each cycle. Price; the conversion information is the advertising cycle, advertising time point, advertising hits, advertising yield, and advertising rate of advertising.
 S5, all advertising features of the advertising bits are mapped to the advertising revenue space to obtain a number of advertising bits feature points of the advertising bits, and perform all the ads of the ad between the advertising bits according to the timing characteristics of the advertising bits. The connection to obtain the timing characteristic trajectory of the advertisement.
 S6, map the advertising feature to the advertisement income space to generate a destination point, and generate an advertising revenue analysis according to the timing characteristic trajectory of the delivery feature point and all advertising bits, and then enter the advertising gain analysis into the advertisement. Advertising trajectory characteristics when dealing with the maximum amount of income, thereby outputting an advertising plant.
 Advertising programs include advertising bits and target ads in the advertising bits of advertising bits, targeted proportion, launch point, and program yields. The advertising list is all the adits involved in the advertising plan; the launch cycle is the time period of the target advertisement in the advertisement, such as February 7 to February 16; the time point is targeted by the target advertisement every day. Time points, such as 6 o'clock in the evening to 12:00; the proportion of transport is the proportion of money spent on each ad between advertising.
 The present invention analyzes the advertising delivery request based on artificial intelligence and big data to determine the advertising laundering program, avoiding the problem of human resources caused by manual planning advertising programs and is not high, and can reduce advertising costs. Improve the benefits of advertising. Furthermore, the present invention is more efficient and higher than that in the prior art manual advertising programs.
 Computer program instructions for performing the operation of the present invention may be assembled instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcodes, firmware instructions, status setting data, or any of one or more programming languages Source code or target code written in combination, the programming language includes object-oriented programming languages, such as SmallTalk, C ++, etc., and process programming languages - such as "C" languages or similar programming languages. Computer readable program instructions can be performed on the user's computer, partially executed on the user's computer, execute as a separate package, partially performed on the remote computer on the remote computer, or on the remote computer or server implement. In the case involving remote computers, remote computers can connect to user computers by any kind of network, including a local area network (LAN) or WAN (WAN), or can be connected to external computers (eg, using Internet service providers through the Internet via the Internet connect). In some embodiments, personalized electronic circuitry, such as a programmable logic circuit, field programmable gate array (FPGA), or programmable logic array (PLA), by using a state information of the computer readable program instruction. Each aspect of the invention is implemented in implementing computer readable program instructions.
 In the above embodiment, it can be achieved through software, hardware, firmware, or any combination thereof in whole or in part. When implemented using software, you can fully or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When loading and executing the computer program instruction on a computer, all or partially generated the flow or function as described in accordance with the present application embodiment. The computer can be a general purpose computer, a dedicated computer, a computer network, or another programmable device. The computer instruction can be stored in a computer readable storage medium, or from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be from one website site, computer, server or data center. Transfer to another website site, computer, server, or data center via wired (such as shaft cable, fiber, digital subscriber) or wireless (eg, infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that the computer can access or a data storage device such as a server, data center, including one or more available media integration. The usable medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape, a photologies (eg, a DVD), or a semiconductor medium (e.g., SolidState Disk, SSD).
 The present application example is described with reference to the flowcharts and / or block diagrams of the method, device (system), and computer program products, in accordance with the embodiment of the present application embodiment. It should be understood that each of the flowcharts and / or blocks in the flowchart and / or block diagram can be implemented by a computer program command, and the binding of the flow and / or box in the flowchart and / or block diagram. These computer program instructions can be provided to a general-purpose computer, a dedicated machine, an embedded processor, or another programmable data processing device to generate a machine such that instructions executed by the processor of the computer or other programmable data processing device. Used to implement the process Figure one Process or multiple processes and / or boxes Figure one Apparatus specified in a plurality of boxes or multiple boxes.
 These computer program instructions can also be stored in a computer readable memory capable of booting a computer or other programmable data processing device in a particular manner, making the instructions stored in the computer readable memory generate a manufacturing product of the instruction device, which Device is implemented in the process Figure one Process or multiple processes and / or boxes Figure one The function specified in the box or multiple boxes.
 It will be apparent to those skilled in the art that various modifications and variations of the present application are made without departing from the spirit and scope of the present application. Thus, the present application is also intended to include these modifications and variations if these expression and variations of the present application embodiments are within the scope of the claims and their equivalents thereof.