Network advertisement release management method and system based on big data

Abstract

The invention discloses a network advertisement release management method and system based on big data, and relates to the technical field of big data, the system comprises a management center, and the management center is connected with a network acquisition module, an intelligent processing module, a management analysis module and a comprehensive adjustment module; user instruction data and advertisement comprehensive data are acquired through a network acquisition module; the intelligent processing module performs image extraction and region inspection on the advertisement comprehensive data to obtain an advertisement feature map; the management analysis module performs morphological transformation on the advertisement feature map to obtain an advertisement feature signal, and sets a regional dynamic coefficient to perform filtering collection and feature filtering on the advertisement feature signal to obtain an advertisement filtering coefficient; performing morphological transformation and feature filtering on the user instruction data in a comprehensive adjustment module to obtain a user filtering coefficient, and performing shielding check on the user filtering coefficient to obtain an updated advertisement coefficient graph and a qualified matching interest spectrum; the accuracy and safety of network advertisement release are improved, and bad advertisements are effectively recognized.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to a network advertisement publishing management method and system based on big data. Background Art

[0002] Big data refers to a collection of data that cannot be captured, managed, and processed by conventional software tools within a certain time frame. Advertising publishing management refers to the process of systematically managing the creativity, production, publishing, monitoring, and effect evaluation of advertisements.

[0003] Due to the limitations of online advertising monitoring technology, inaccurate monitoring may occur, resulting in inaccurate evaluation of advertising effectiveness. The online advertisements released to users are not of interest to the users, or the users are not very interested in them, and cannot provide users with a reasonable advertising experience. Therefore, it is of great theoretical and practical significance to conduct interest matching and type blocking before online advertisements are released.

[0004] How to use big data technology to extract images from the collected comprehensive advertising data, obtain advertising feature maps, convert the advertising feature maps into advertising feature signals for feature filtering, obtain advertising filtering coefficients, perform morphological transformation on user instruction data, obtain user filtering coefficients, shield and check the advertising filtering coefficients according to the user filtering coefficients, and obtain updated advertising coefficient maps and qualified matching interest spectra are the problems we need to solve. To this end, a network advertising publishing management method and system based on big data are provided. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for managing online advertising publishing based on big data.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The network advertising publishing management system based on big data includes a management center, which is connected to a network acquisition module, an intelligent processing module, a management analysis module and a comprehensive adjustment module;

[0008] The process of the network acquisition module collecting user instruction data and advertisement comprehensive data includes:

[0009] Set up the user end and the collection end;

[0010] Obtain the target user by performing identity registration and account login on the user end;

[0011] The user terminal issues an extraction instruction to the target user, and the target user collects instructions according to the received extraction instruction to obtain user instruction data;

[0012] Collect information about online advertisements through the collection terminal to obtain comprehensive advertising data.

[0013] The process of obtaining an advertisement feature map includes:

[0014] Performing image extraction on the obtained comprehensive advertisement data to obtain a comprehensive advertisement image;

[0015] Setting a cropping size, and performing a standard check on the advertisement comprehensive image according to the obtained cropping size to obtain a standard advertisement image;

[0016] Set the screening pixel base, and set the feature acquisition window according to the obtained screening pixel base;

[0017] The advertisement comprehensive image is inspected in regions according to the obtained feature acquisition window to obtain an advertisement feature map.

[0018] The process of obtaining the control level includes:

[0019] Performing morphological transformation on the obtained advertisement feature graph to obtain an advertisement feature signal;

[0020] Extracting characteristics of the obtained advertisement characteristic signal to obtain characteristic parameters;

[0021] The filtering coefficient is set according to the obtained characteristic parameters, and the obtained filtering coefficient is dynamically designed to obtain a form processing end;

[0022] Set the regional dynamic coefficient, and set the control parameters according to the obtained regional dynamic coefficient;

[0023] The regional dynamic coefficient is divided into intervals according to the obtained control parameters to obtain the control level.

[0024] The process of obtaining the advertisement filtering coefficient includes:

[0025] According to the obtained control series, the regional dynamic coefficient is segmented to obtain the control regional coefficient segment;

[0026] The advertisement characteristic signal is filtered and collected according to the obtained control area coefficient segment to obtain the filtering control coefficient, and the obtained filtering control coefficient is uploaded to the form processing end, and the advertisement characteristic signal is characteristically filtered through the form processing end to obtain the advertisement filtering coefficient.

[0027] The process of obtaining the forbidden coefficient difference includes:

[0028] Performing a morphological transformation on the obtained user command data to obtain a user command signal;

[0029] Perform feature filtering on the user command signal through the form processing end to obtain the user filtering coefficient;

[0030] The user filtering coefficients include an interest filtering coefficient and a prohibited filtering coefficient;

[0031] Generate an interest coefficient map according to the obtained interest filtering coefficient, generate a prohibition coefficient map according to the obtained prohibition filtering coefficient, and generate an advertisement coefficient map according to the obtained advertisement filtering coefficient;

[0032] The advertisement coefficient map is overlapped and matched according to the prohibited coefficient map to obtain the prohibited coefficient difference.

[0033] The process of obtaining an updated advertising coefficient graph and a qualified matching interest spectrum includes:

[0034] Setting a preference threshold, performing a threshold comparison on the forbidden coefficient difference according to the preference threshold, and obtaining a matching forbidden spectrum;

[0035] Perform shielding check on the advertisement coefficient graph according to the obtained matching forbidden spectrum to obtain an updated advertisement coefficient graph;

[0036] Perform overlap matching on the updated advertisement coefficient graph according to the obtained interest coefficient graph to obtain a deviation coefficient value;

[0037] The deviation coefficient value is threshold matched according to the obtained preference threshold to obtain a qualified matching interest spectrum.

[0038] Based on the above-mentioned network advertisement publishing management system based on big data, the present invention also provides a network advertisement publishing management method based on big data, comprising the following steps:

[0039] Step 1: Collect user instruction data and advertising comprehensive data;

[0040] Step 2: extract the image of the comprehensive advertisement data to obtain the comprehensive advertisement image, set the cropping size to perform standard verification on the comprehensive advertisement image to obtain the standard advertisement image, set the feature collection window, perform regional verification on the comprehensive advertisement image according to the feature collection window to obtain the advertisement feature map;

[0041] Step 3: Perform morphological transformation and feature extraction on the advertisement feature map to obtain feature parameters, perform dynamic design based on the feature parameters, obtain a form processing end, set control parameters to divide the regional dynamic coefficient into intervals, obtain a control level, perform level segmentation on the regional dynamic coefficient based on the control level, obtain a control regional coefficient segment, filter and collect the advertisement feature signal based on the control regional coefficient segment, obtain a filtering control coefficient and upload it to the form processing end, perform feature filtering on the advertisement feature signal through the form processing end, and obtain an advertisement filtering coefficient;

[0042] Step 4: Perform morphological transformation and feature filtering on the user command data to obtain the user filtering coefficient, generate an interest coefficient map and a prohibited coefficient map based on the user filtering coefficient, generate an advertising coefficient map based on the advertising filtering coefficient, perform overlap matching on the advertising coefficient map based on the prohibited coefficient map to obtain the prohibited coefficient difference, perform shielding check and threshold matching on the advertising coefficient map through the prohibited coefficient difference to obtain an updated advertising coefficient map and a qualified matching interest spectrum.

[0043] Compared with the prior art, the invention has the following beneficial effects: performing image extraction and region inspection on the collected comprehensive advertising data, obtaining an advertising feature map and performing morphological transformation, obtaining an advertising feature signal, setting a regional dynamic coefficient to filter and collect the advertising feature signal and filter the feature, and obtaining an advertising filtering coefficient; converting the collected network advertising data from different sources into the same form of advertising filtering coefficients, reducing the complexity of network advertising data management, and improving data quality;

[0044] The user instruction data is transformed in shape and feature filtered to obtain the user filtering coefficient, and the advertisement filtering coefficient is screened and checked and threshold matched through the user filtering coefficient to obtain an updated advertisement coefficient graph and a qualified matching interest spectrum; by improving the matching degree of the network advertisements that the user is interested in, increasing the publishing accuracy, improving the effect of network advertisements, and screening the types of advertisements prohibited by the user, it is beneficial to improve the quality of advertisements, conduct safe delivery, and conduct more accurate monitoring of network advertisements. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0047] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0048] like Figure 1 As shown, the network advertising publishing management system based on big data includes a management center, and the management center is connected to a network acquisition module, an intelligent processing module, a management analysis module and a comprehensive adjustment module;

[0049] The network collection module is used to collect user instruction data and advertisement comprehensive data, and the specific process includes:

[0050] Set up the user end and the collection end;

[0051] Register the user's identity through the user terminal and obtain the login account and password;

[0052] Furthermore, the identity registration means that the user's personal information and login password are inputted at the user end, the personal information is reviewed by the management center, and a login account is generated for the user who passes the review according to the personal information;

[0053] Log in the user account according to the obtained login account and login password to obtain the target user;

[0054] Furthermore, the account login means that the user's login account and login password are input into the user terminal to log in, and the user who successfully logs in is marked as the target user; the target user means an authenticated network user who can receive the target object of the network advertisement;

[0055] The user terminal issues an extraction instruction to the target user, and the target user collects instructions according to the received extraction instruction to obtain user instruction data, wherein the user instruction data includes user interest data and user prohibited data;

[0056] Further, the extraction instruction indicates that a user advertising survey is conducted on the target user before the online advertisement is delivered, so that the online advertisement delivered to the target user is of interest to the user and does not contain user-forbidden content, and the user-forbidden content indicates online advertisements with related content selected by the user that are prohibited from being delivered to the target user terminal; for example, the user interest data of the target user is food and travel, and the user-forbidden data is shopping and education and training;

[0057] Collect information about online advertisements through the collection terminal to obtain comprehensive advertisement data;

[0058] It should be further explained that, in the specific implementation process, the collected comprehensive advertising data are online advertisements that are prepared to be released to target users, including banner ads, text link ads, pop-up ads, video ads, and search engine ads.

[0059] The intelligent processing module is used to extract images and check regions of the comprehensive advertisement data to obtain advertisement feature maps. The specific process includes:

[0060] Performing image extraction on the obtained comprehensive advertisement data to obtain a comprehensive advertisement image;

[0061] Further, the image extraction means performing image collection on the obtained advertising comprehensive data according to different representation forms of the advertising comprehensive data, that is, collecting pictures corresponding to each advertising comprehensive data, and statistically analyzing the advertising comprehensive images corresponding to each network advertisement, that is, a plurality of images can be extracted from one advertising comprehensive data, but the extracted plurality of images all represent the same network advertisement, that is, the network advertisement corresponding to the advertising comprehensive data for which the image extraction is performed;

[0062] Setting a cropping size, wherein the cropping size includes image width, image length, and recognition criteria;

[0063] Furthermore, the image length represents the number of pixels in the vertical direction, and the image width represents the number of pixels in the horizontal direction; the recognition criteria represent the colors that are recognized and allowed to pass, and if a color is not within the recognition criteria, the color cannot pass after the cropping size is calibrated;

[0064] Performing standard calibration on the advertisement comprehensive image according to the obtained cropping size to obtain a standard advertisement image;

[0065] Perform pixel recognition on standard advertising images to obtain advertising pixel points;

[0066] Furthermore, the advertisement pixel point represents the specific position of the pixel in the standard advertisement image;

[0067] Setting a screening pixel base, wherein the screening pixel base is expressed in a function form;

[0068] Setting a feature acquisition window according to the obtained screening pixel base number;

[0069] The feature acquisition window is a matrix composed of a number of elements, and the obtained screening pixel base is uploaded to each element in the feature acquisition window;

[0070] Performing a regional check on the advertisement comprehensive image according to the obtained feature acquisition window to obtain an advertisement feature map;

[0071] It should be further explained that, in the specific implementation process, the process of regional inspection includes:

[0072] Setting a check starting point in the advertisement comprehensive image, wherein the check starting point means selecting any point in the advertisement comprehensive image as the starting point of the area check. In this embodiment, the upper left corner of the advertisement comprehensive image is used as the check starting point;

[0073] Uploading the obtained feature acquisition window to the inspection starting point of the advertisement comprehensive image, and marking the corresponding position of the feature acquisition window in the advertisement comprehensive image as a window area;

[0074] Perform regional convolution on the window area according to the obtained feature acquisition window to obtain a regional advertising image;

[0075] Furthermore, the regional convolution means convolving the elements in the feature acquisition window with the advertising pixel points at the corresponding position of the window area, that is, convolving the screening pixel base with the advertising pixel points at the corresponding position to obtain a regional advertising image;

[0076] The feature acquisition window is translated and slid in the advertising comprehensive image to reach the next window area, and the feature acquisition window and the window area are regionally convolved until the advertising comprehensive image is covered by the feature acquisition window. The obtained regional advertising images are counted, and the obtained regional advertising images are sequentially combined according to the obtained advertising comprehensive image to obtain an advertising feature map.

[0077] The management and analysis module is used to perform morphological transformation on the advertisement feature map, obtain advertisement feature signals, set regional dynamic coefficients to filter and collect advertisement feature signals and filter features, and obtain advertisement filtering coefficients. The specific process includes:

[0078] Performing morphological transformation on the obtained advertisement feature graph to obtain an advertisement feature signal;

[0079] Furthermore, the morphological transformation represents converting the advertisement feature graph into a signal form;

[0080] Extracting characteristics of the obtained advertisement characteristic signal to obtain characteristic parameters;

[0081] Further, the characteristic parameters include signal bandwidth, frequency content, noise level, signal duration, signal rise and fall time;

[0082] Setting a filtering coefficient according to the obtained characteristic parameters, wherein the filtering coefficient includes a cutoff frequency, a roll-off filter, a bandwidth parameter, and a filtering order;

[0083] It should be further explained that, in the specific implementation process, the filtering coefficient is determined according to the characteristic parameters of the advertisement characteristic signal;

[0084] Dynamically design the obtained filtering coefficient to obtain the form processing end;

[0085] Furthermore, the dynamic design means uploading the obtained filtering coefficients to the filter design software, and generating the corresponding filter by inputting the filtering coefficients, i.e., the formal processing end;

[0086] Setting a regional dynamic coefficient, wherein the regional dynamic coefficient is expressed in a function form;

[0087] Setting control parameters according to the obtained regional dynamic coefficients, wherein the control parameters are used to adjust parameters of scaling and translation transformation of the regional dynamic coefficients in the time dimension and the frequency dimension;

[0088] According to the obtained control parameters, the regional dynamic coefficient is divided into intervals to obtain the control level;

[0089] It should be further explained that, in the specific implementation process, the interval division process includes:

[0090] The distance between two adjacent control parameters is obtained, recorded as the parameter interval, and the control level is obtained according to the obtained parameter interval, the control parameter and the advertising characteristic signal, and the obtained control level is marked as J, where: L represents the signal length of the advertising characteristic signal, k represents the parameter interval, m represents the number of parameter intervals, c represents the control parameter, B represents the signal bandwidth, and f0 represents the cutoff frequency;

[0091] According to the obtained control series, the regional dynamic coefficient is segmented to obtain the control regional coefficient segment;

[0092] Furthermore, the stage segmentation means dividing the regional dynamic coefficient into control regional coefficient segments of equal length according to the obtained control stage;

[0093] The advertisement characteristic signal is filtered and collected according to the obtained control area coefficient segment to obtain the filtering control coefficient;

[0094] Furthermore, the process of filtering and collecting includes:

[0095] The obtained control region coefficient segments are sequentially uploaded to the advertising characteristic signal, and the control region coefficient is convolved with the corresponding position of the advertising characteristic signal to obtain the control coefficient segment, and the control coefficient segment is combined according to the obtained advertising characteristic signal to obtain the filtered control coefficient;

[0096] Upload the obtained filtering control coefficient to the form processing end;

[0097] Perform feature filtering on the advertisement feature signal through the form processing end to obtain an advertisement filtering coefficient;

[0098] It should be further explained that, in the specific implementation process, the feature filtering process includes:

[0099] The processing end performs filtering response on the advertisement characteristic signal according to the obtained form to obtain the response advertisement coefficient, convolves the obtained filtering control coefficient with the response advertisement coefficient to obtain the filtering convolution coefficient, performs inverse discrete cosine transform on the obtained filtering convolution coefficient to obtain the advertisement filtering coefficient.

[0100] The comprehensive adjustment module is used to perform morphological transformation and feature filtering on the user instruction data, obtain the user filtering coefficient, perform shielding inspection on the user filtering coefficient, obtain the updated advertisement coefficient map and the qualified matching interest spectrum, and the specific process includes:

[0101] Performing a morphological transformation on the obtained user command data to obtain a user command signal;

[0102] Perform feature filtering on the user command signal through the form processing end to obtain the user filtering coefficient;

[0103] The user filtering coefficients include an interest filtering coefficient and a prohibited filtering coefficient;

[0104] Generate an interest coefficient map according to the obtained interest filtering coefficient, and mark the obtained interest coefficient map with a curve to obtain an interest coefficient spectrum line;

[0105] Further, the curve mark represents a frequency spectrum curve generated according to the interest filtering coefficient in the interest coefficient map;

[0106] Generate a prohibited coefficient graph according to the obtained prohibited filtering coefficients, and mark the obtained prohibited coefficient graph with a curve to obtain a prohibited coefficient spectrum line;

[0107] It should be further explained that, in the specific implementation process, the generated interest coefficient graph includes the interest coefficient spectra generated by all user interest data of the target user, that is, the interest coefficient spectra obtained according to the user interest data of the target user are in the same interest coefficient graph; similarly, the prohibited coefficient graph includes all prohibited coefficient spectra generated according to the user prohibited data, and one target user can only generate one interest coefficient graph and one prohibited coefficient graph;

[0108] Generate an advertising coefficient graph according to the obtained advertising filtering coefficient, and mark the obtained advertising coefficient graph with a curve to obtain an advertising coefficient spectrum line;

[0109] Performing overlap matching on the advertisement coefficient map according to the obtained prohibited coefficient map to obtain a prohibited coefficient difference;

[0110] Furthermore, the overlap matching process includes:

[0111] Select any advertising coefficient spectrum line in the advertising coefficient graph and record it as the reference advertising line;

[0112] According to the obtained reference advertising line, the prohibited coefficient spectrum lines in the prohibited coefficient graph are differentially marked to obtain the deviation coefficient value, and the obtained deviation coefficient value is marked as P, wherein P = ∫ [R (t) - Q (t)] dt, R (t) represents the reference advertising line, Q (t) represents the prohibited coefficient spectrum line, t represents the coordinate node of the horizontal axis in the advertising coefficient graph and the prohibited coefficient graph, and dt represents the differential of t; until all the prohibited coefficient spectrum lines in the prohibited coefficient graph are differentially marked with the reference advertising line;

[0113] Remaining advertising coefficient spectrum lines are replaced until each advertising coefficient spectrum line in the advertising coefficient graph is recorded as a reference advertising line;

[0114] Setting a preference threshold, performing threshold comparison on the forbidden coefficient difference according to the obtained preference threshold, and obtaining a matching forbidden spectrum and an irrelevant forbidden spectrum;

[0115] Further, the advertising coefficient spectrum line corresponding to the forbidden coefficient difference value less than the preference threshold is marked as a matching forbidden spectrum, that is, the network advertisement is a forbidden content for the target user, and the network advertisement is intercepted and blocked by the management center, and is not released to the target user; the advertising coefficient spectrum line corresponding to the forbidden coefficient difference value greater than or equal to the preference threshold is marked as an irrelevant forbidden spectrum, and the network advertisement is retained for interest verification;

[0116] Perform shielding check on the advertisement coefficient graph according to the obtained matching forbidden spectrum to obtain an updated advertisement coefficient graph;

[0117] Furthermore, the shielding check means deleting the advertising coefficient spectrum line corresponding to the matching prohibited spectrum in the advertising coefficient graph, that is, removing the network advertisements matching the user prohibited data of the target user, until there is no advertising coefficient spectrum line corresponding to the matching prohibited spectrum in the advertising coefficient graph, and obtaining an updated advertising coefficient graph;

[0118] Perform overlap matching on the updated advertisement coefficient graph according to the obtained interest coefficient graph to obtain a deviation coefficient value;

[0119] Perform threshold matching on the deviation coefficient value according to the obtained preference threshold to obtain a qualified matching interest spectrum and an irrelevant matching interest spectrum;

[0120] It should be further explained that, in the specific implementation process, the interest coefficient spectrum corresponding to the deviation coefficient value less than the preference threshold is marked as a qualified matching interest spectrum, that is, the pushed network advertisement meets the interest of the target user; the interest coefficient spectrum corresponding to the deviation coefficient value greater than or equal to the preference threshold is marked as an irrelevant matching interest spectrum, that is, the pushed network advertisement is not the content of interest to the target user;

[0121] Furthermore, the number of qualified matching interest spectra of the same online advertisement is counted to obtain the number of qualified spectra, and the obtained number of qualified spectra is sorted in descending order to obtain a qualified interest push sequence, and interest advertisements are pushed to target users according to the qualified interest push sequence; in particular, the obtained qualified interest push sequence is an advertisement coefficient spectrum line sorted according to the number of qualified spectra.

[0122] Based on the above-mentioned network advertisement publishing management system based on big data, the present invention also provides a network advertisement publishing management method based on big data, comprising the following steps:

[0123] Step 1: Collect user instruction data and advertising comprehensive data;

[0124] Step 2: extract the image of the comprehensive advertisement data to obtain the comprehensive advertisement image, set the cropping size to perform standard verification on the comprehensive advertisement image to obtain the standard advertisement image, set the feature collection window, perform regional verification on the comprehensive advertisement image according to the feature collection window to obtain the advertisement feature map;

[0125] Step 3: Perform morphological transformation and feature extraction on the advertisement feature map to obtain feature parameters, perform dynamic design based on the feature parameters, obtain a form processing end, set control parameters to divide the regional dynamic coefficient into intervals, obtain a control level, perform level segmentation on the regional dynamic coefficient based on the control level, obtain a control regional coefficient segment, filter and collect the advertisement feature signal based on the control regional coefficient segment, obtain a filtering control coefficient and upload it to the form processing end, perform feature filtering on the advertisement feature signal through the form processing end, and obtain an advertisement filtering coefficient;

[0126] Step 4: Perform morphological transformation and feature filtering on the user command data to obtain the user filtering coefficient, generate an interest coefficient map and a prohibited coefficient map based on the user filtering coefficient, generate an advertising coefficient map based on the advertising filtering coefficient, perform overlap matching on the advertising coefficient map based on the prohibited coefficient map to obtain the prohibited coefficient difference, perform shielding check and threshold matching on the advertising coefficient map through the prohibited coefficient difference to obtain an updated advertising coefficient map and a qualified matching interest spectrum.

[0127] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A network advertising publishing management system based on big data, including a management center, characterized in that: The management center is connected to a network acquisition module, an intelligent processing module, a management analysis module and a comprehensive adjustment module; The network collection module is used to collect user instruction data and advertisement comprehensive data; The intelligent processing module is used to extract images from the comprehensive advertising data to obtain a comprehensive advertising image, set a cropping size to perform standard verification on the comprehensive advertising image to obtain a standard advertising image, set a feature collection window, perform regional verification on the comprehensive advertising image according to the feature collection window, and obtain an advertising feature map; The management and analysis module is used to perform morphological transformation and feature extraction on the advertisement feature graph, obtain feature parameters, perform dynamic design according to the feature parameters, obtain a form processing end, set control parameters to divide the regional dynamic coefficient into intervals, obtain a control level, perform level segmentation on the regional dynamic coefficient according to the control level, obtain a control regional coefficient segment, filter and collect the advertisement feature signal according to the control regional coefficient segment, obtain a filtering control coefficient and upload it to the form processing end, perform feature filtering on the advertisement feature signal through the form processing end, and obtain an advertisement filtering coefficient; The comprehensive adjustment module is used to perform morphological transformation and feature filtering on user command data, obtain user filtering coefficients, generate interest coefficient maps and prohibited coefficient maps according to the user filtering coefficients, generate advertising coefficient maps according to the advertising filtering coefficients, perform overlap matching on the advertising coefficient maps according to the prohibited coefficient maps, obtain prohibited coefficient differences, perform shielding checks and threshold matching on the advertising coefficient maps through the prohibited coefficient differences, and obtain updated advertising coefficient maps and qualified matching interest spectra.

2. The network advertising publishing management system based on big data according to claim 1 is characterized in that: The process of the network acquisition module collecting user instruction data and advertisement comprehensive data includes: Set up the user end and the collection end; Obtain the target user by performing identity registration and account login on the user end; The user terminal issues an extraction instruction to the target user, and the target user collects instructions according to the received extraction instruction to obtain user instruction data; Collect information about online advertisements through the collection terminal to obtain comprehensive advertising data.

3. The network advertising publishing management system based on big data according to claim 2 is characterized in that: The process of obtaining an advertisement feature map includes: Performing image extraction on the obtained comprehensive advertisement data to obtain a comprehensive advertisement image; Setting a cropping size, and performing a standard check on the advertisement comprehensive image according to the obtained cropping size to obtain a standard advertisement image; Set the screening pixel base, and set the feature acquisition window according to the obtained screening pixel base; The advertisement comprehensive image is inspected in regions according to the obtained feature acquisition window to obtain an advertisement feature map.

4. The network advertising publishing management system based on big data according to claim 3 is characterized in that: The process of obtaining the control level includes: Performing morphological transformation on the obtained advertisement feature graph to obtain an advertisement feature signal; Extracting characteristics of the obtained advertisement characteristic signal to obtain characteristic parameters; The filtering coefficient is set according to the obtained characteristic parameters, and the obtained filtering coefficient is dynamically designed to obtain a form processing end; Set the regional dynamic coefficient, and set the control parameters according to the obtained regional dynamic coefficient; The regional dynamic coefficient is divided into intervals according to the obtained control parameters to obtain the control level.

5. The network advertising publishing management system based on big data according to claim 4 is characterized in that: The process of obtaining the advertisement filtering coefficient includes: According to the obtained control series, the regional dynamic coefficient is segmented to obtain the control regional coefficient segment; The advertisement characteristic signal is filtered and collected according to the obtained control area coefficient segment to obtain the filtering control coefficient, and the obtained filtering control coefficient is uploaded to the form processing end, and the advertisement characteristic signal is characteristically filtered through the form processing end to obtain the advertisement filtering coefficient.

6. The network advertising publishing management system based on big data according to claim 5 is characterized in that: The process of obtaining the forbidden coefficient difference includes: Performing a morphological transformation on the obtained user command data to obtain a user command signal; Perform feature filtering on the user command signal through the form processing end to obtain the user filtering coefficient; The user filtering coefficients include an interest filtering coefficient and a prohibited filtering coefficient; Generate an interest coefficient map according to the obtained interest filtering coefficient, generate a prohibition coefficient map according to the obtained prohibition filtering coefficient, and generate an advertisement coefficient map according to the obtained advertisement filtering coefficient; The advertisement coefficient map is overlapped and matched according to the prohibited coefficient map to obtain the prohibited coefficient difference.

7. The network advertising publishing management system based on big data according to claim 6 is characterized in that: The process of obtaining an updated advertising coefficient graph and a qualified matching interest spectrum includes: Setting a preference threshold, performing a threshold comparison on the forbidden coefficient difference according to the preference threshold, and obtaining a matching forbidden spectrum; Perform shielding check on the advertisement coefficient graph according to the obtained matching forbidden spectrum to obtain an updated advertisement coefficient graph; Perform overlap matching on the updated advertisement coefficient graph according to the obtained interest coefficient graph to obtain a deviation coefficient value; The deviation coefficient value is threshold matched according to the obtained preference threshold to obtain a qualified matching interest spectrum.

8. The network advertisement publishing management method based on the network advertisement publishing management system of big data according to any one of claims 1 to 7, characterized in that: The following steps are involved: Step 1: Collect user instruction data and advertising comprehensive data; Step 2: extract the image of the comprehensive advertisement data to obtain the comprehensive advertisement image, set the cropping size to perform standard verification on the comprehensive advertisement image to obtain the standard advertisement image, set the feature collection window, perform regional verification on the comprehensive advertisement image according to the feature collection window to obtain the advertisement feature map; Step 3: Perform morphological transformation and feature extraction on the advertisement feature map to obtain feature parameters, perform dynamic design based on the feature parameters, obtain a form processing end, set control parameters to divide the regional dynamic coefficient into intervals, obtain a control level, perform level segmentation on the regional dynamic coefficient based on the control level, obtain a control regional coefficient segment, filter and collect the advertisement feature signal based on the control regional coefficient segment, obtain a filtering control coefficient and upload it to the form processing end, perform feature filtering on the advertisement feature signal through the form processing end, and obtain an advertisement filtering coefficient; Step 4: Perform morphological transformation and feature filtering on the user command data to obtain the user filtering coefficient, generate an interest coefficient map and a prohibited coefficient map based on the user filtering coefficient, generate an advertising coefficient map based on the advertising filtering coefficient, perform overlap matching on the advertising coefficient map based on the prohibited coefficient map to obtain the prohibited coefficient difference, perform shielding check and threshold matching on the advertising coefficient map through the prohibited coefficient difference to obtain an updated advertising coefficient map and a qualified matching interest spectrum.

Images