Cross-platform advertisement pushing and feedback integration system and method based on cloud computing

Abstract

The invention discloses a cross-platform advertisement pushing and feedback integration system and method based on cloud computing, and relates to the technical field of cloud computing, and the method comprises the steps: constructing a cloud database to carry out advertisement information synchronous updating on each platform, and constructing a classification information form to carry out overall planning, integration and classification management on the advertisement information; dividing period periods, and carrying out activity index evaluation on each advertisement of the corresponding type in different period periods on each platform; analyzing the actual browsing rate of each advertisement of the corresponding type in different period periods based on the evaluation data in combination with the corresponding screening model; an actual effect feature matrix corresponding to each advertisement in each platform is constructed, and platform user advertisement association degree analysis is performed on each advertisement in different period periods in combination with the real-time browsing rate of each advertisement in different period periods and historical data of the same type of advertisements in the same platform; and based on the platform user advertisement correlation analysis data of each advertisement in the corresponding platform in the corresponding period period, making a period resource allocation scheduling plan for each platform advertisement data and outputting the plan.

Description

Technical Field

[0001] The present invention relates to the field of cloud computing, and in particular to a cross-platform advertising push and feedback integration system and method based on cloud computing. Background Art

[0002] The cross-platform advertising push and feedback integration system based on cloud computing is a comprehensive solution that uses the elastic computing and storage capabilities of cloud services to achieve intelligent distribution and precise positioning of advertising content across multiple platforms, and analyzes and optimizes based on user feedback data, aiming to improve advertising efficiency. At present, Internet information can achieve information synchronization and spread rapidly over a large area, so most information transmission platforms have come into being, which has also brought new ways of advertising dissemination. Broadcasting and transmission through real-time information synchronization platforms can efficiently and quickly carry out advertising and consume resources; however, the delivery of a large amount of advertising resources over a large area and in a short period of time will make it impossible to deliver targeted advertising resources according to the characteristics of different platforms, which can easily lead to a waste of advertising resources. Summary of the invention

[0003] The purpose of the present invention is to provide a cross-platform advertising push and feedback integration system and method based on cloud computing to solve the problems raised in the prior art.

[0004] To achieve the above object, the present invention provides the following technical solutions: A cross-platform advertising push and feedback integration method based on cloud computing, the method comprising the following steps: S100, synchronously updating the advertising information of each platform by building a cloud advertising data access database, and building a classified information form to coordinate, integrate and classify the advertising information; S200, by dividing the periodic time periods and combining the classified information form, the activity index of each type of advertisement in different periodic time periods of each platform is evaluated; based on the evaluation data combined with the corresponding screening model, the effective browsing rate of each type of advertisement in different periodic time periods is analyzed; S300, by constructing an effectiveness feature matrix corresponding to each advertisement on each platform, combining the real-time browsing rate of each advertisement of the corresponding type on each platform in different periodic periods with the historical data of the same type of advertisements on the same platform, the platform user advertisement correlation analysis is performed on each advertisement of the corresponding type on each platform in different periodic periods; S400: Based on the platform user advertising relevance analysis data of each advertisement in the corresponding platform during the corresponding periodic period, a time period resource allocation scheduling plan is performed for each platform advertising data and a scheduling planning plan is output.

[0005] The specific steps of S100 for synchronously updating the advertising information of each platform by constructing a cloud advertising data access database and constructing a classified information form to coordinate, integrate and classify the advertising information are as follows: S101, by building a cloud advertising data access database, the advertising delivery platform is associated with data, and the advertising delivery information and user browsing advertising information in each platform are synchronously obtained and updated and stored; S102. Integrate the advertising delivery information of each platform and the information on user browsing of advertisements retrieved from the cloud-based advertising data access database, and classify and process the information according to the advertising attribute data; construct a classified information form based on the integrated and classified advertising information; the advertising delivery information includes the platform information corresponding to each advertisement, the advertisement type information, the number of advertisements delivered, etc.; the user browsing of advertisements includes the user's corresponding platform identity number, the user's browsing of advertisements, etc.; the advertising attribute data corresponds to the platform information and the advertisement type information of each advertisement; the user browsing of advertisements includes likes, favorites, forwarding, comments, etc.

[0006] The S200 evaluates the activity index of each advertisement of the corresponding type in different periodic periods on each platform by dividing the periodic periods and combining the classified information form; the specific steps of analyzing the effective browsing rate of each advertisement of the corresponding type in different periodic periods based on the evaluation data and the corresponding screening model are as follows: S201, divide the whole day into periods, divide the periodic periods into equal time length specifications and mark each periodic period with a number; retrieve the advertising information from the classified information form in the cloud advertising data access database corresponding to each periodic period; evaluate the activity index of each platform in the corresponding periodic period according to the advertising information of each platform in the corresponding periodic period, and the calculation formula is: ; Wherein, Ac(i,j,t) is the advertising activity index of the advertisement of the corresponding type number j in the corresponding platform number i during the corresponding period of time t; m(i,j) is the average number of active users in the corresponding platform number i during the corresponding period of time t; M(i) is the number of registered users in the corresponding platform number i; n(i,j,t) is the number of advertisements of the corresponding type number j in the corresponding platform number i during the corresponding period of time t; N(i,t) is the total number of advertisements in the corresponding platform number i during the corresponding period of time t; i is the platform number; j is the advertisement type number; t is the period number; S202, based on the activity index evaluation data of the corresponding types of advertisements on each platform in the corresponding periodic period, respectively analyze the effective browsing rate of the corresponding types of advertisements; retrieve the browsing operation information of the users on the platform in the corresponding periodic period for a certain advertisement; build a screening model for screening and marking the user's advertisement browsing operation; if the user has a browsing operation on a certain advertisement, the advertisement is recorded as a valid browsing record; if the user has no browsing operation on a certain advertisement, then by setting an effective browsing time threshold, if the user's browsing time is greater than or equal to the threshold, it is recorded as a valid browsing record; then based on the effective browsing data of each advertisement in the corresponding type on each platform in different periodic periods combined with the activity index of the corresponding advertisement type on the corresponding platform in the corresponding periodic period, the effective browsing rate analysis is performed, and its calculation formula is: ; Wherein, Ar(i,j,k,t) is the effective browsing rate of the advertisement numbered k in the corresponding type numbered j in the corresponding platform numbered i during the corresponding period of time numbered t; y(i,j,k,t) is the number of valid browsing records of the advertisement numbered k in the corresponding type numbered j in the corresponding platform numbered i during the corresponding period of time numbered t; Y(i,j,k,t) is the total number of browsing records of the advertisement numbered k in the corresponding type numbered j in the corresponding platform numbered i during the corresponding period of time numbered t; the total number of browsing records includes both the number of valid records and the number of invalid records where the user has no browsing operation on a certain advertisement and the browsing time is less than a threshold.

[0007] The specific steps of S300 in performing platform user advertisement correlation analysis on advertisements of the corresponding types on each platform in different periodic periods by constructing an effectiveness feature matrix corresponding to each advertisement on each platform, combining the real-time browsing rate of advertisements of the corresponding types on each platform in different periodic periods with the historical data of advertisements of the same type on the same platform are as follows: S301. During the corresponding period, construct an effective feature matrix for each advertisement of the corresponding type on each platform; take the total number of browsing records of a certain advertisement in the corresponding period as the total number of matrix elements, mark the corresponding matrix element position of the valid browsing record corresponding to the advertisement as 1, and mark the corresponding element positions as 0, and then complete the construction of the effective feature matrix of the corresponding advertisement in the corresponding platform during the corresponding period; retrieve the user browsing data of the advertisement with the maximum active index in the historical advertising data on the corresponding platform during the period where the corresponding advertisement is located, and construct the corresponding effective feature matrix, which is recorded as the historical best advertisement effective feature matrix of the corresponding platform during the corresponding period; wherein the element position marked with 0 corresponds to the invalid browsing record; S302, combining the effectiveness feature matrix of a certain advertisement on a certain platform in the current cycle period and the historical best advertisement effectiveness feature matrix of the corresponding platform in the current cycle period, analyzing the correlation between the advertisements of the platform users in the cycle period of the current advertisement; performing matrix corresponding feature vector analysis on the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, respectively, and obtaining the feature vectors corresponding to the effectiveness feature matrix of the corresponding target advertisement and the historical best advertisement effectiveness feature matrix, respectively; extracting the elements at the same position of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, and constructing an associated feature vector; the associated feature vector is extracted by extracting the elements at the same position of the effectiveness feature matrix of the target advertisement and the historical best advertisement effectiveness feature matrix, and mapping them into a coordinate vector constructed by pointing one element of the two elements to another element; based on the feature vector processing of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, analyzing the correlation between the advertisements of the platform users in the cycle period of the current advertisement, and the calculation formula is: ; Among them, Rl(i,j,k,t) is the platform user advertising relevance of the advertisement numbered k in the corresponding type numbered j in the corresponding platform numbered i in the corresponding periodic period numbered t; Ar(Hmax) is the maximum value of the historical advertising activity index of the corresponding target advertisement in the same periodic period and on the same platform; Z(p,q) is the associated feature vector of the corresponding element position (p,q) of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix; p*q is the number of matrix elements; W(k) is the feature vector corresponding to the effectiveness feature matrix of the current target advertisement; W(H) is the feature vector corresponding to the historical best advertisement effectiveness feature matrix; According to the platform user advertising relevance analysis data of the current advertising cycle, a judgment threshold Rl(t) is introduced; if Rl(i,j,k,t)<Rl(t), it is judged that the relevance of the current advertisement in the corresponding platform during the current cycle is low, and a prompt to stop advertising is marked; if Rl(i,j,k,t)≥Rl(t), it is judged that the relevance of the current advertisement in the corresponding platform during the current cycle is normal, and normal advertising is marked.

[0008] The specific steps of performing time period resource allocation scheduling planning for each platform advertisement data and outputting the scheduling planning scheme based on the platform user advertisement association analysis data of each advertisement in the corresponding platform during the corresponding periodic period in S400 are as follows: S401. Based on the platform user advertising relevance analysis data of each advertisement in the corresponding platform during the corresponding period, the platform user advertising relevance between different advertisements in the same platform and the platform user advertising relevance between the same advertisements on different platforms are obtained; by performing a proportion analysis on the platform user advertising relevance between different advertisements in the same platform and the platform user advertising relevance between the same advertisements on different platforms, the proportion of the number of different advertisements delivered in the same platform is adjusted and an advertising delivery plan is output; S402: Regulate cross-platform advertising resources through the output advertising plan.

[0009] A cross-platform advertising push and feedback integration system based on cloud computing, the system includes an advertising data acquisition module, an advertising data evaluation module, a matrix correlation analysis module and a planning and control module; The advertising data acquisition module synchronously updates the advertising information of each platform by constructing a cloud-based advertising data access database, and constructs a classified information form to coordinate, integrate and classify the advertising information; the advertising data evaluation module evaluates the activity index of each type of advertisement in different periodic periods on each platform by dividing the periodic time period and combining the classified information form; based on the evaluation data and combined with the corresponding screening model, the effective browsing rate of each type of advertisement in different periodic time periods is analyzed; the matrix association analysis module constructs an effective feature matrix corresponding to each advertisement in each platform, combines the real-time browsing rate of each type of advertisement in different periodic time periods on each platform and the historical data of the same type of advertisements on the same platform, and analyzes the platform user advertising correlation of each type of advertisement on each platform in different periodic time periods; the planning and control module performs time period resource allocation scheduling planning for each platform advertising data based on the platform user advertising correlation analysis data of each advertisement on the corresponding platform in the corresponding periodic time period and outputs the scheduling planning plan.

[0010] The advertisement data acquisition module includes an advertisement data acquisition unit and a data integration processing unit; The advertising data acquisition unit associates data with the advertising delivery platforms by building a cloud advertising data access database, and synchronously acquires, updates and stores the advertising delivery information and user browsing advertising information on each platform; The data integration processing unit comprehensively integrates the advertising information of each platform and the user browsing advertising information retrieved from the cloud advertising data access database, and classifies and processes the information according to the advertising attribute data; and constructs a classified information form based on the comprehensively integrated and classified advertising information.

[0011] The advertising data evaluation module includes an activity index evaluation unit and an effective browsing rate analysis unit; The activity index evaluation unit divides the whole day into periods, divides the periodic periods into equal time length specifications and numbers each periodic period; retrieves the advertising delivery information from the classified information form in the cloud advertising data access database corresponding to each periodic period; and evaluates the activity index of each platform in the corresponding periodic period according to the advertising delivery information of each platform in the corresponding periodic period; The effective browsing rate analysis unit analyzes the effective browsing rates of the corresponding advertisements of each type based on the activity index evaluation data of the corresponding type of advertisements on each platform within the corresponding cycle period; retrieves the browsing operation information of the users on the platform within the corresponding period for a certain advertisement; builds a screening model for screening and marking the user's advertisement browsing operations; if the user has a browsing operation on a certain advertisement, the advertisement is recorded as a valid browsing record; if the user has no browsing operation on a certain advertisement, an effective browsing time threshold is set, and if the user's browsing time is greater than or equal to the threshold, it is recorded as a valid browsing record; then, based on the effective browsing data of each advertisement of the corresponding type on each platform within different cycle periods and the activity index of the corresponding advertisement type on the corresponding platform within the corresponding cycle period, an effective browsing rate analysis is performed.

[0012] The matrix association analysis module includes an effective feature matrix construction unit and a platform user association analysis unit; The effective feature matrix construction unit constructs the effective feature matrix for each advertisement of the corresponding type in each platform in the corresponding period; the total number of browsing records of a certain advertisement in the corresponding period is taken as the total number of matrix elements, and the corresponding matrix element position of the valid browsing record corresponding to the advertisement is marked as 1, and the corresponding element positions are marked as 0, thereby completing the construction of the effective feature matrix of the corresponding advertisement in the corresponding platform in the corresponding period; retrieve the user browsing data of the advertisement with the maximum value of the activity index in the historical advertisement data in the corresponding platform in the period where the corresponding advertisement is located, and construct the corresponding effective feature matrix, which is recorded as the historical best advertisement effective feature matrix of the corresponding platform in the corresponding period; The platform user association analysis unit combines the effectiveness feature matrix of an advertisement on a platform in the current cycle period with the historical best advertisement effectiveness feature matrix of the corresponding platform in the current cycle period to analyze the platform user advertisement association of the current advertisement; performs matrix corresponding feature vector analysis on the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, respectively, and obtains feature vectors corresponding to the effectiveness feature matrix of the corresponding target advertisement and the historical best advertisement effectiveness feature matrix, respectively; extracts the same-position elements of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, and constructs an association feature vector; performs platform user advertisement association analysis on the cycle period of the current advertisement based on feature vector processing of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix; and performs association judgment on advertisements on each platform in the current cycle period based on the platform user advertisement association analysis data of the cycle period of the current advertisement, by introducing a judgment threshold.

[0013] The planning and control module includes an advertising control planning unit and a planning output unit; The advertising control and planning unit obtains the platform user advertising correlation between different advertisements in the same platform and the platform user advertising correlation between the same advertisements on different platforms based on the platform user advertising correlation analysis data of each advertisement in the corresponding platform during the corresponding period; controls the proportion of the number of different advertisements on the same platform and outputs the advertising delivery plan by performing a proportion analysis on the platform user advertising correlation between different advertisements on the same platform and the platform user advertising correlation between the same advertisements on different platforms; The planning output unit performs cross-platform advertising delivery resource regulation through the output advertising delivery plan.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention integrates the advertising data of multiple platforms by constructing a cross-platform cloud service database, and synchronously updates the data information of the advertisements of each platform; constructs an advertising information form based on a series of classification processing based on the advertising data information of each platform, performs time period activity index analysis on the advertising data of each platform based on the form, and performs effective browsing rate analysis of the corresponding advertisements in combination with the user's advertising browsing operation; constructs an effective feature matrix for each advertisement in combination with the user's browsing operation, and performs correlation analysis in combination with historical data to obtain the platform user advertising correlation of each advertisement in the corresponding time period; makes platform advertising screening recommendations based on the correlation analysis data, and gives a corresponding cross-platform resource allocation plan for the advertisement to realize advertising resource regulation; the present invention realizes data performance analysis of cross-platform advertisements through series advertising data analysis; based on this, obtains the correlation between different advertisements of the corresponding advertising resources on different platforms or on the same platform, effectively improves the waste of advertising resources, and realizes the precise delivery of advertising resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a structural schematic diagram of the cross-platform advertising push and feedback integration system based on cloud computing of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Example: Figure 1 As shown, the present invention provides a technical solution: A cross-platform advertising push and feedback integration method based on cloud computing, the method comprising the following steps: S100, synchronously updating the advertising information of each platform by building a cloud advertising data access database, and building a classified information form to coordinate, integrate and classify the advertising information; S200, by dividing the periodic time periods and combining the classified information form, the activity index of each type of advertisement in different periodic time periods of each platform is evaluated; based on the evaluation data combined with the corresponding screening model, the effective browsing rate of each type of advertisement in different periodic time periods is analyzed; S300, by constructing an effectiveness feature matrix corresponding to each advertisement on each platform, combining the real-time browsing rate of each advertisement of the corresponding type on each platform in different periodic periods with the historical data of the same type of advertisements on the same platform, the platform user advertisement correlation analysis is performed on each advertisement of the corresponding type on each platform in different periodic periods; S400: Based on the platform user advertising relevance analysis data of each advertisement in the corresponding platform during the corresponding periodic period, a time period resource allocation scheduling plan is performed for each platform advertising data and a scheduling planning plan is output.

[0018] The specific steps of S100 for synchronously updating the advertising information of each platform by constructing a cloud advertising data access database and constructing a classified information form to coordinate, integrate and classify the advertising information are as follows: S101, by building a cloud advertising data access database, the advertising delivery platform is associated with data, and the advertising delivery information and user browsing advertising information in each platform are synchronously obtained and updated and stored; S102: The advertising information of each platform and the information of user browsing advertisements retrieved from the cloud-based advertising data access database are integrated and classified according to the advertisement attribute data; and a classified information form is constructed according to the integrated and classified advertisement information.

[0019] The S200 evaluates the activity index of each advertisement of the corresponding type in different periodic periods on each platform by dividing the periodic periods and combining the classified information form; the specific steps of analyzing the effective browsing rate of each advertisement of the corresponding type in different periodic periods based on the evaluation data and the corresponding screening model are as follows: S201, divide the whole day into periods, divide the periodic periods into equal time length specifications and mark each periodic period with a number; retrieve the advertising information from the classified information form in the cloud advertising data access database corresponding to each periodic period; evaluate the activity index of each platform in the corresponding periodic period according to the advertising information of each platform in the corresponding periodic period, and the calculation formula is: ; Wherein, Ac(i,j,t) is the advertising activity index of the advertisement of the corresponding type number j in the corresponding platform number i during the corresponding period of time t; m(i,j) is the average number of active users in the corresponding platform number i during the corresponding period of time t; M(i) is the number of registered users in the corresponding platform number i; n(i,j,t) is the number of advertisements of the corresponding type number j in the corresponding platform number i during the corresponding period of time t; N(i,t) is the total number of advertisements in the corresponding platform number i during the corresponding period of time t; i is the platform number; j is the advertisement type number; t is the period number; S202, based on the activity index evaluation data of the corresponding types of advertisements on each platform in the corresponding periodic period, respectively analyze the effective browsing rate of the corresponding types of advertisements; retrieve the browsing operation information of the users on the platform in the corresponding periodic period for a certain advertisement; build a screening model for screening and marking the user's advertisement browsing operation; if the user has a browsing operation on a certain advertisement, the advertisement is recorded as a valid browsing record; if the user has no browsing operation on a certain advertisement, then by setting an effective browsing time threshold, if the user's browsing time is greater than or equal to the threshold, it is recorded as a valid browsing record; then based on the effective browsing data of each advertisement in the corresponding type on each platform in different periodic periods combined with the activity index of the corresponding advertisement type on the corresponding platform in the corresponding periodic period, the effective browsing rate analysis is performed, and its calculation formula is: ; Wherein, Ar(i,j,k,t) is the effective browsing rate of the advertisement numbered k in the corresponding type numbered j on the corresponding platform i during the corresponding period of time t; y(i,j,k,t) is the number of effective browsing records of the advertisement numbered k in the corresponding type numbered j on the corresponding platform i during the corresponding period of time t; Y(i,j,k,t) is the total number of browsing records of the advertisement numbered k in the corresponding type numbered j on the corresponding platform i during the corresponding period of time t.

[0020] The specific steps of S300 in performing platform user advertisement correlation analysis on advertisements of the corresponding types on each platform in different periodic periods by constructing an effectiveness feature matrix corresponding to each advertisement on each platform, combining the real-time browsing rate of advertisements of the corresponding types on each platform in different periodic periods with the historical data of advertisements of the same type on the same platform are as follows: S301. During the corresponding period, construct an effective feature matrix for each advertisement of the corresponding type on each platform; take the total number of browsing records of a certain advertisement in the corresponding period as the total number of matrix elements, mark the corresponding matrix element position of the valid browsing record corresponding to the advertisement as 1, and mark the corresponding element positions as 0, and then complete the construction of the effective feature matrix of the corresponding advertisement in the corresponding platform during the corresponding period; retrieve the user browsing data of the advertisement with the maximum value of the activity index in the historical advertisement data on the corresponding platform during the period in which the corresponding advertisement is located, and construct the corresponding effective feature matrix, which is recorded as the historical best advertisement effective feature matrix of the corresponding platform during the corresponding period; S302, combining the effectiveness feature matrix of a certain advertisement on a certain platform in the current cycle period and the historical best advertisement effectiveness feature matrix of the corresponding platform in the current cycle period, analyzing the platform user advertisement association of the current advertisement; performing matrix corresponding feature vector analysis on the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, respectively, obtaining the feature vectors corresponding to the effectiveness feature matrix of the corresponding target advertisement and the historical best advertisement effectiveness feature matrix, respectively; extracting the same-position elements of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, and constructing the association feature vector; based on the feature vector processing of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, analyzing the platform user advertisement association of the current advertisement cycle period, the calculation formula is: ; Among them, Rl(i,j,k,t) is the platform user advertising relevance of the advertisement numbered k in the corresponding type numbered j in the corresponding platform numbered i in the corresponding periodic period numbered t; Ar(Hmax) is the maximum value of the historical advertising activity index of the corresponding target advertisement in the same periodic period and on the same platform; Z(p,q) is the associated feature vector of the corresponding element position (p,q) of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix; p*q is the number of matrix elements; W(k) is the feature vector corresponding to the effectiveness feature matrix of the current target advertisement; W(H) is the feature vector corresponding to the historical best advertisement effectiveness feature matrix; According to the platform user advertising relevance analysis data of the current advertising cycle, a judgment threshold Rl(t) is introduced; if Rl(i,j,k,t)<Rl(t), it is judged that the relevance of the current advertisement in the corresponding platform during the current cycle is low, and a prompt to stop advertising is marked; if Rl(i,j,k,t)≥Rl(t), it is judged that the relevance of the current advertisement in the corresponding platform during the current cycle is normal, and normal advertising is marked.

[0021] The specific steps of performing time period resource allocation scheduling planning for each platform advertisement data and outputting the scheduling planning scheme based on the platform user advertisement association analysis data of each advertisement in the corresponding platform during the corresponding periodic period in S400 are as follows: S401. Based on the platform user advertising relevance analysis data of each advertisement in the corresponding platform during the corresponding period, the platform user advertising relevance between different advertisements in the same platform and the platform user advertising relevance between the same advertisements on different platforms are obtained; by performing a proportion analysis on the platform user advertising relevance between different advertisements in the same platform and the platform user advertising relevance between the same advertisements on different platforms, the proportion of the number of different advertisements delivered in the same platform is adjusted and an advertising delivery plan is output; S402: Regulate cross-platform advertising resources through the output advertising plan.

[0022] A cross-platform advertising push and feedback integration system based on cloud computing, the system includes an advertising data acquisition module, an advertising data evaluation module, a matrix correlation analysis module and a planning and control module; The advertising data acquisition module synchronously updates the advertising information of each platform by constructing a cloud-based advertising data access database, and constructs a classified information form to coordinate, integrate and classify the advertising information; the advertising data evaluation module evaluates the activity index of each type of advertisement in different periodic periods on each platform by dividing the periodic time period and combining the classified information form; based on the evaluation data and combined with the corresponding screening model, the effective browsing rate of each type of advertisement in different periodic time periods is analyzed; the matrix association analysis module constructs an effective feature matrix corresponding to each advertisement in each platform, combines the real-time browsing rate of each type of advertisement in different periodic time periods on each platform and the historical data of the same type of advertisements on the same platform, and analyzes the platform user advertising correlation of each type of advertisement on each platform in different periodic time periods; the planning and control module performs time period resource allocation scheduling planning for each platform advertising data based on the platform user advertising correlation analysis data of each advertisement on the corresponding platform in the corresponding periodic time period and outputs the scheduling planning plan.

[0023] The advertisement data acquisition module includes an advertisement data acquisition unit and a data integration processing unit; The advertising data acquisition unit associates data with the advertising delivery platforms by building a cloud advertising data access database, and synchronously acquires, updates and stores the advertising delivery information and user browsing advertising information on each platform; The data integration processing unit comprehensively integrates the advertising information of each platform and the user browsing advertising information retrieved from the cloud advertising data access database, and classifies and processes the information according to the advertising attribute data; and constructs a classified information form based on the comprehensively integrated and classified advertising information.

[0024] The advertising data evaluation module includes an activity index evaluation unit and an effective browsing rate analysis unit; The activity index evaluation unit divides the whole day into periods, divides the periodic periods into equal time length specifications and numbers each periodic period; retrieves the advertising delivery information from the classified information form in the cloud advertising data access database corresponding to each periodic period; and evaluates the activity index of each platform in the corresponding periodic period according to the advertising delivery information of each platform in the corresponding periodic period; The effective browsing rate analysis unit analyzes the effective browsing rates of the corresponding advertisements of each type based on the activity index evaluation data of the corresponding type of advertisements on each platform within the corresponding cycle period; retrieves the browsing operation information of the users on the platform within the corresponding period for a certain advertisement; builds a screening model for screening and marking the user's advertisement browsing operations; if the user has a browsing operation on a certain advertisement, the advertisement is recorded as a valid browsing record; if the user has no browsing operation on a certain advertisement, an effective browsing time threshold is set, and if the user's browsing time is greater than or equal to the threshold, it is recorded as a valid browsing record; then, based on the effective browsing data of each advertisement of the corresponding type on each platform within different cycle periods and the activity index of the corresponding advertisement type on the corresponding platform within the corresponding cycle period, an effective browsing rate analysis is performed.

[0025] The matrix association analysis module includes an effective feature matrix construction unit and a platform user association analysis unit; The effective feature matrix construction unit constructs the effective feature matrix for each advertisement of the corresponding type in each platform in the corresponding period; the total number of browsing records of a certain advertisement in the corresponding period is taken as the total number of matrix elements, and the corresponding matrix element position of the valid browsing record corresponding to the advertisement is marked as 1, and the corresponding element positions are marked as 0, thereby completing the construction of the effective feature matrix of the corresponding advertisement in the corresponding platform in the corresponding period; retrieve the user browsing data of the advertisement with the maximum value of the activity index in the historical advertisement data in the corresponding platform in the period where the corresponding advertisement is located, and construct the corresponding effective feature matrix, which is recorded as the historical best advertisement effective feature matrix of the corresponding platform in the corresponding period; The platform user association analysis unit combines the effectiveness feature matrix of an advertisement on a platform in the current cycle period with the historical best advertisement effectiveness feature matrix of the corresponding platform in the current cycle period to analyze the platform user advertisement association of the current advertisement; performs matrix corresponding feature vector analysis on the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, respectively, and obtains feature vectors corresponding to the effectiveness feature matrix of the corresponding target advertisement and the historical best advertisement effectiveness feature matrix, respectively; extracts the same-position elements of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, and constructs an association feature vector; performs platform user advertisement association analysis on the cycle period of the current advertisement based on feature vector processing of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix; and performs association judgment on advertisements on each platform in the current cycle period based on the platform user advertisement association analysis data of the cycle period of the current advertisement, by introducing a judgment threshold.

[0026] The planning and control module includes an advertising control planning unit and a planning output unit; The advertising control and planning unit obtains the platform user advertising correlation between different advertisements in the same platform and the platform user advertising correlation between the same advertisements on different platforms based on the platform user advertising correlation analysis data of each advertisement in the corresponding platform during the corresponding period; controls the proportion of the number of different advertisements on the same platform and outputs the advertising delivery plan by performing a proportion analysis on the platform user advertising correlation between different advertisements on the same platform and the platform user advertising correlation between the same advertisements on different platforms; The planning output unit performs cross-platform advertising delivery resource regulation through the output advertising delivery plan; In the examples: Now an advertising company needs to monitor its advertising resources across platforms, so it adopts the cross-platform advertising push and feedback integration system based on cloud computing of the present invention; it builds a cloud advertising data access database to associate data with the advertising delivery platform, synchronously obtains and updates and stores the advertising delivery information and user browsing advertising information on each platform; it integrates the advertising delivery information and user browsing advertising information on each platform retrieved from the cloud advertising data access database, and classifies and processes them according to the advertising attribute data; it builds a classified information form according to the integrated and classified advertising information; By dividing the whole day into periods, taking equal periodic periods to divide them into equal time length specifications and marking each period with numbers; respectively retrieve the advertising information from the classified information form in the cloud advertising data access database corresponding to each periodic period; and evaluate the activity index of each platform in the corresponding periodic period according to the advertising information of each platform in the corresponding periodic period. The calculation formula is: ; Based on the activity index evaluation data of the corresponding types of advertisements on each platform in the corresponding period, the effective browsing rate of the corresponding advertisements of each type is analyzed respectively; combined with the browsing operation information of the users on the platform in the corresponding period, the browsing operation information of a certain advertisement is retrieved; a screening model is constructed to screen and mark the user's advertising browsing operation; if the user has a browsing operation on a certain advertisement, the advertisement is recorded as a valid browsing record; if the user has no browsing operation on a certain advertisement, an effective browsing time threshold is set, and if the user's browsing time is greater than or equal to the threshold, it is recorded as a valid browsing record; then based on the effective browsing data of each advertisement of the corresponding type on each platform in different periodic periods, combined with the activity index of the corresponding advertisement type on the corresponding platform in the corresponding periodic period, the effective browsing rate analysis is performed, and its calculation formula is: ; In the corresponding period, construct the effectiveness feature matrix for each advertisement of the corresponding type in each platform; take the total number of browsing records of a certain advertisement in the corresponding period as the total number of matrix elements, mark the corresponding matrix element position of the valid browsing record of the advertisement as 1, and mark the corresponding element positions as 0, and then complete the construction of the effectiveness feature matrix of the corresponding advertisement in the corresponding platform in the corresponding period; retrieve the user browsing data of the advertisement with the maximum activity index in the historical advertising data in the corresponding platform in the period where the corresponding advertisement is located, and construct the corresponding effectiveness feature matrix, which is recorded as the historical best advertisement effectiveness feature matrix of the corresponding platform in the corresponding period; Combine the effectiveness feature matrix of a certain advertisement on a certain platform in the current cycle period with the historical best advertisement effectiveness feature matrix of the corresponding platform in the current cycle period to analyze the platform user advertisement association of the current advertisement; perform matrix corresponding feature vector analysis on the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix respectively, and obtain the feature vectors corresponding to the effectiveness feature matrix of the corresponding target advertisement and the historical best advertisement effectiveness feature matrix respectively; extract the elements at the same position of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, and construct the association feature vector; based on the feature vector processing of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, analyze the platform user advertisement association of the current advertisement cycle period, and the calculation formula is: ; According to the platform user advertising relevance analysis data of the current advertising cycle, the judgment threshold Rl(t) is introduced; if Rl(i,j,k,t)<Rl(t), it is judged that the relevance of the current advertisement in the corresponding platform in the current cycle is low, and a prompt is marked to suggest stopping advertising delivery; if Rl(i,j,k,t)≥Rl(t), it is judged that the relevance of the current advertisement in the corresponding platform in the current cycle is normal, and normal advertising delivery is marked; Based on the platform user advertising association analysis data of each advertisement in the corresponding platform during the corresponding period, the platform user advertising association between different advertisements in the same platform and the platform user advertising association between the same advertisements on different platforms are obtained; by performing proportional analysis on the platform user advertising association between different advertisements in the same platform and the platform user advertising association between the same advertisements on different platforms, the proportion of the number of different advertisements on the same platform is regulated and an advertising delivery plan is output; cross-platform advertising delivery resources are regulated through the output advertising delivery plan.

[0027] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A cross-platform advertising push and feedback integration method based on cloud computing, characterized by: The method comprises the following steps: S100, synchronously updating the advertising information of each platform by building a cloud advertising data access database, and building a classified information form to coordinate, integrate and classify the advertising information; S200, by dividing the periodic time periods and combining the classified information form, the activity index of each type of advertisement in different periodic time periods of each platform is evaluated; based on the evaluation data combined with the corresponding screening model, the effective browsing rate of each type of advertisement in different periodic time periods is analyzed; S300, by constructing an effectiveness feature matrix corresponding to each advertisement on each platform, combining the real-time browsing rate of each advertisement of the corresponding type on each platform in different periodic periods with the historical data of the same type of advertisements on the same platform, the platform user advertisement correlation analysis is performed on each advertisement of the corresponding type on each platform in different periodic periods; S400: Based on the platform user advertising relevance analysis data of each advertisement in the corresponding platform during the corresponding periodic period, a time period resource allocation scheduling plan is performed for each platform advertising data and a scheduling planning plan is output.

2. The cross-platform advertising push and feedback integration method based on cloud computing according to claim 1 is characterized by: The specific steps of S100 for synchronously updating the advertising information of each platform by constructing a cloud advertising data access database and constructing a classified information form to coordinate, integrate and classify the advertising information are as follows: S101, by building a cloud advertising data access database, the advertising delivery platform is associated with data, and the advertising delivery information and user browsing advertising information in each platform are synchronously obtained and updated and stored; S102: The advertising information of each platform and the information of user browsing advertisements retrieved from the cloud-based advertising data access database are integrated and classified according to the advertisement attribute data; and a classified information form is constructed according to the integrated and classified advertisement information.

3. The cross-platform advertising push and feedback integration method based on cloud computing according to claim 2 is characterized by: The S200 evaluates the activity index of each advertisement of the corresponding type in different periodic periods on each platform by dividing the periodic periods and combining the classified information form; the specific steps of analyzing the effective browsing rate of each advertisement of the corresponding type in different periodic periods based on the evaluation data and the corresponding screening model are as follows: S201, divide the whole day into periods, divide the periodic periods into equal time length specifications and mark each periodic period with a number; retrieve the advertising information from the classified information form in the cloud advertising data access database corresponding to each periodic period; evaluate the activity index of each platform in the corresponding periodic period according to the advertising information of each platform in the corresponding periodic period, and the calculation formula is: ； Wherein, Ac(i,j,t) is the advertising activity index of the advertisement of the corresponding type number j in the corresponding platform number i during the corresponding period of time t; m(i,j) is the average number of active users in the corresponding platform number i during the corresponding period of time t; M(i) is the number of registered users in the corresponding platform number i; n(i,j,t) is the number of advertisements of the corresponding type number j in the corresponding platform number i during the corresponding period of time t; N(i,t) is the total number of advertisements in the corresponding platform number i during the corresponding period of time t; i is the platform number; j is the advertisement type number; t is the period number; S202, based on the activity index evaluation data of the corresponding types of advertisements on each platform in the corresponding periodic period, respectively analyze the effective browsing rate of the corresponding types of advertisements; retrieve the browsing operation information of the users on the platform in the corresponding periodic period for a certain advertisement; build a screening model for screening and marking the user's advertisement browsing operation; if the user has a browsing operation on a certain advertisement, the advertisement is recorded as a valid browsing record; if the user has no browsing operation on a certain advertisement, then by setting an effective browsing time threshold, if the user's browsing time is greater than or equal to the threshold, it is recorded as a valid browsing record; then based on the effective browsing data of each advertisement in the corresponding type on each platform in different periodic periods combined with the activity index of the corresponding advertisement type on the corresponding platform in the corresponding periodic period, the effective browsing rate analysis is performed, and its calculation formula is: ； Wherein, Ar(i,j,k,t) is the effective browsing rate of the advertisement numbered k in the corresponding type numbered j on the corresponding platform i during the corresponding period of time t; y(i,j,k,t) is the number of effective browsing records of the advertisement numbered k in the corresponding type numbered j on the corresponding platform i during the corresponding period of time t; Y(i,j,k,t) is the total number of browsing records of the advertisement numbered k in the corresponding type numbered j on the corresponding platform i during the corresponding period of time t.

4. The cross-platform advertising push and feedback integration method based on cloud computing according to claim 3 is characterized by: The specific steps of S300 in performing platform user advertisement correlation analysis on advertisements of the corresponding types on each platform in different periodic periods by constructing an effectiveness feature matrix corresponding to each advertisement on each platform, combining the real-time browsing rate of advertisements of the corresponding types on each platform in different periodic periods with the historical data of advertisements of the same type on the same platform are as follows: S301, constructing an effectiveness feature matrix for each advertisement of a corresponding type on each platform during a corresponding period; The total number of browsing records of a certain advertisement in the corresponding period is taken as the total number of matrix elements, and the corresponding matrix element position of the valid browsing record corresponding to the advertisement is marked as 1, and the corresponding element positions are marked as 0, so as to complete the construction of the effective feature matrix of the corresponding advertisement in the corresponding platform in the corresponding period; retrieve the user browsing data of the advertisement with the maximum activity index in the historical advertising data on the corresponding platform in the period where the corresponding advertisement is located, and construct the corresponding effective feature matrix, which is recorded as the historical best advertising effective feature matrix of the corresponding platform in the corresponding period; S302, combining the effectiveness feature matrix of a certain advertisement on a certain platform in the current cycle period and the historical best advertisement effectiveness feature matrix of the corresponding platform in the current cycle period, analyzing the platform user advertisement association of the current advertisement; performing matrix corresponding feature vector analysis on the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, respectively, obtaining the feature vectors corresponding to the effectiveness feature matrix of the corresponding target advertisement and the historical best advertisement effectiveness feature matrix, respectively; extracting the same-position elements of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, and constructing the association feature vector; based on the feature vector processing of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, analyzing the platform user advertisement association of the current advertisement cycle period, the calculation formula is: ； Among them, Rl(i,j,k,t) is the platform user advertising relevance of the advertisement numbered k in the corresponding type numbered j in the corresponding platform numbered i in the corresponding periodic period numbered t; Ar(Hmax) is the maximum value of the historical advertising activity index of the corresponding target advertisement in the same periodic period and on the same platform; Z(p,q) is the associated feature vector of the corresponding element position (p,q) of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix; p*q is the number of matrix elements; W(k) is the feature vector corresponding to the effectiveness feature matrix of the current target advertisement; W(H) is the feature vector corresponding to the historical best advertisement effectiveness feature matrix; According to the platform user advertising relevance analysis data of the current advertising cycle, a judgment threshold Rl(t) is introduced; if Rl(i,j,k,t)<Rl(t), it is judged that the relevance of the current advertisement in the corresponding platform during the current cycle is low, and a prompt to stop advertising is marked; if Rl(i,j,k,t)≥Rl(t), it is judged that the relevance of the current advertisement in the corresponding platform during the current cycle is normal, and normal advertising is marked.

5. The cross-platform advertising push and feedback integration method based on cloud computing according to claim 4 is characterized by: The specific steps of performing time period resource allocation scheduling planning for each platform advertisement data and outputting the scheduling planning scheme based on the platform user advertisement association analysis data of each advertisement in the corresponding platform during the corresponding periodic period in S400 are as follows: S401. Based on the platform user advertising relevance analysis data of each advertisement in the corresponding platform during the corresponding period, the platform user advertising relevance between different advertisements in the same platform and the platform user advertising relevance between the same advertisements on different platforms are obtained; by performing a proportion analysis on the platform user advertising relevance between different advertisements in the same platform and the platform user advertising relevance between the same advertisements on different platforms, the proportion of the number of different advertisements delivered in the same platform is adjusted and an advertising delivery plan is output; S402: Regulate cross-platform advertising resources through the output advertising plan.

6. A cross-platform advertising push and feedback integration system based on cloud computing, characterized by: The system includes an advertising data acquisition module, an advertising data evaluation module, a matrix association analysis module and a planning and control module; The advertising data acquisition module synchronously updates the advertising information of each platform by constructing a cloud-based advertising data access database, and constructs a classified information form to coordinate, integrate and classify the advertising information; the advertising data evaluation module evaluates the activity index of each type of advertisement in different periodic periods on each platform by dividing the periodic time period and combining the classified information form; based on the evaluation data and combined with the corresponding screening model, the effective browsing rate of each type of advertisement in different periodic time periods is analyzed; the matrix association analysis module constructs an effective feature matrix corresponding to each advertisement in each platform, combines the real-time browsing rate of each type of advertisement in different periodic time periods on each platform and the historical data of the same type of advertisements on the same platform, and analyzes the platform user advertising correlation of each type of advertisement on each platform in different periodic time periods; the planning and control module performs time period resource allocation scheduling planning for each platform advertising data based on the platform user advertising correlation analysis data of each advertisement on the corresponding platform in the corresponding periodic time period and outputs the scheduling planning plan.

7. The cross-platform advertising push and feedback integration system based on cloud computing according to claim 6, characterized in that: The advertisement data acquisition module includes an advertisement data acquisition unit and a data integration processing unit; The advertising data acquisition unit associates data with the advertising delivery platforms by building a cloud advertising data access database, and synchronously acquires, updates and stores the advertising delivery information and user browsing advertising information on each platform; The data integration processing unit comprehensively integrates the advertising information of each platform and the user browsing advertising information retrieved from the cloud advertising data access database, and classifies and processes the information according to the advertising attribute data; and constructs a classified information form based on the comprehensively integrated and classified advertising information.

8. The cloud computing-based cross-platform advertising push and feedback integration system according to claim 7, characterized in that: The advertising data evaluation module includes an activity index evaluation unit and an effective browsing rate analysis unit; The activity index evaluation unit divides the whole day into periods, divides the periodic periods into equal time length specifications and numbers each periodic period; retrieves the advertising delivery information from the classified information form in the cloud advertising data access database corresponding to each periodic period; and evaluates the activity index of each platform in the corresponding periodic period according to the advertising delivery information of each platform in the corresponding periodic period; The effective browsing rate analysis unit analyzes the effective browsing rate of each type of corresponding advertisement based on the activity index evaluation data of the corresponding type of advertisements on each platform in the corresponding period; retrieves the browsing operation information of a certain advertisement by the user on the platform in the corresponding period; and constructs a screening model for screening and marking the user's advertisement browsing operation; If a user browses an advertisement, it will be recorded as a valid browsing record. If a user does not browse an advertisement, an effective browsing time threshold is set. If the user's browsing time is greater than or equal to the threshold, it will be recorded as a valid browsing record. Based on the effective browsing data of each advertisement of the corresponding type on each platform in different cycle time periods, combined with the activity index of the corresponding advertisement type on the corresponding platform in the corresponding cycle time period, an effective browsing rate analysis is performed.

9. The cloud computing-based cross-platform advertising push and feedback integration system according to claim 8, characterized in that: The matrix association analysis module includes an effective feature matrix construction unit and a platform user association analysis unit; The effective feature matrix construction unit constructs the effective feature matrix for each advertisement of the corresponding type on each platform in the corresponding period; The total number of browsing records of a certain advertisement in the corresponding period is taken as the total number of matrix elements, and the corresponding matrix element position of the valid browsing record corresponding to the advertisement is marked as 1, and the corresponding element positions are marked as 0, so as to complete the construction of the effective feature matrix of the corresponding advertisement in the corresponding platform in the corresponding period; retrieve the user browsing data of the advertisement with the maximum activity index in the historical advertising data on the corresponding platform in the period where the corresponding advertisement is located, and construct the corresponding effective feature matrix, which is recorded as the historical best advertising effective feature matrix of the corresponding platform in the corresponding period; The platform user association analysis unit combines the effectiveness feature matrix of an advertisement on a platform in the current cycle period with the historical best advertisement effectiveness feature matrix of the corresponding platform in the current cycle period to analyze the platform user advertisement association of the current advertisement; performs matrix corresponding feature vector analysis on the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, respectively, and obtains feature vectors corresponding to the effectiveness feature matrix of the corresponding target advertisement and the historical best advertisement effectiveness feature matrix, respectively; extracts the same-position elements of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix, and constructs an association feature vector; performs platform user advertisement association analysis on the cycle period of the current advertisement based on feature vector processing of the effectiveness feature matrix of the current target advertisement and the historical best advertisement effectiveness feature matrix; and performs association judgment on advertisements on each platform in the current cycle period based on the platform user advertisement association analysis data of the cycle period of the current advertisement, by introducing a judgment threshold.

10. The cross-platform advertising push and feedback integration system based on cloud computing according to claim 9, characterized in that: The planning and control module includes an advertising control planning unit and a planning output unit; The advertising control and planning unit obtains the platform user advertising correlation between different advertisements in the same platform and the platform user advertising correlation between the same advertisements on different platforms based on the platform user advertising correlation analysis data of each advertisement in the corresponding platform during the corresponding period; controls the proportion of the number of different advertisements on the same platform and outputs the advertising delivery plan by performing a proportion analysis on the platform user advertising correlation between different advertisements on the same platform and the platform user advertising correlation between the same advertisements on different platforms; The planning output unit performs cross-platform advertising delivery resource regulation through the output advertising delivery plan.

Images