Digital advertisement putting method, system and equipment based on big data analysis

Abstract

The invention relates to the technical field of digital advertisement putting, in particular to a digital advertisement putting method, system and equipment based on big data analysis, comprising the following steps: step 1, collecting various user behavior data related to advertisement content, step 2, analyzing the various user behavior data of the digital advertisement putting method collected in the step 1, 3, based on the keyword analysis result in the step 2, analyzing the relevancy between the keyword and the content of the target digital advertisement, and 4, based on the relevancy in the step 3 and the user behavior data in the step 1, analyzing the content of the target digital advertisement. Step 5, performing evaluation based on a digital advertisement putting result in the step 4, and step 6, performing man-machine interaction based on the evaluation of the digital advertisement putting result in the step 5. Through a big data analysis technology, personalized recommendation and accurate putting of the advertisement content are realized, and then the efficiency and effect of advertisement putting are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of digital advertising delivery, and in particular to a digital advertising delivery method, system and device based on big data analysis. Background Art

[0002] Big data analysis refers to the analysis of large-scale data. Data analysis, data mining and other commercial value-oriented applications of big data have gradually become the focus of profit that industry professionals are vying for. With the rapid development of the Internet, digital advertising has become an important means for companies to promote their products and services. Big data analysis technology provides data decision support for the precise placement of digital advertising.

[0003] In the field of digital advertising, traditional digital advertising methods mainly rely on limited data support, rough user portraits, static bidding strategies and simple matching mechanisms. Advertisers or advertising agencies will manually select a series of keywords related to products or services based on the content of the advertisement. The selection of these keywords is mainly based on experience and industry knowledge to ensure that the advertisement can reach potential users who may be interested. Traditional methods collect and analyze limited user data and try to build user portraits to understand the basic characteristics and interest preferences of the target audience.

[0004] However, there are still some shortcomings in its actual use. For example, the effect of digital advertising is largely affected by keywords. Accurate keywords can increase the exposure and click-through rate of advertisements. However, traditional keyword selection often relies on manual experience and is difficult to adapt to the rapidly changing market environment. In order to solve these problems, precise delivery methods based on big data analysis have emerged. By collecting and analyzing user behavior data, keywords are dynamically optimized to achieve personalized recommendations and precise delivery of advertising content, thereby improving the efficiency and effectiveness of advertising delivery. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a digital advertisement delivery method based on big data analysis to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: a digital advertising delivery method based on big data analysis, comprising the following steps:

[0007] Step 1: Collect various user behavior data related to advertising content through log analysis and user behavior tracking technology, and build a database for analyzing digital advertising delivery methods;

[0008] Step 2: Analyze various user behavior data of the digital advertising delivery method collected in step 1 to obtain keyword analysis results of digital advertising delivery;

[0009] Step 3: Based on the keyword analysis results in step 2, the relevance between the keywords and the target digital advertisement content is analyzed by natural language processing technology to obtain a keyword library related to the target digital advertisement product, and generate a second keyword list;

[0010] Step 4: Based on the second keyword list in step 3 and the user behavior data in step 1, digital advertisement content is matched in real time, and digital advertisements are delivered to target user groups;

[0011] Step 5: Evaluate the digital advertising delivery results in step 4 to obtain digital advertising delivery effect indicators;

[0012] Step 6: Conduct human-computer interaction on the digital advertising effect indicators obtained in step 5.

[0013] Preferably, the user behavior data in step 2 include user search keyword data, user browsing record data, user purchase record data, and competitor advertising data, wherein the competitor advertising data includes the keywords used and the bidding strategies.

[0014] Preferably, the keyword analysis result in step 2 includes the following steps:

[0015] Step 2.1: Place the data processing unit:

[0016] Step 2.1.1: Data preprocessing: Through data preprocessing technology, the data quality index Dqi is obtained. Where Normalization() represents the data normalization function, Ad represents abnormal data, Id represents invalid data, and Td represents total data;

[0017] Step 2.1.2: Keyword screening: Screen the user search keywords related to the advertising content from the search log, and preliminarily generate the first keyword list Kt(1), Kt(1) = [Kt1, Kt2, Kt i ,...,Kt n ], where Kt i represents the ith keyword related to the target advertising content, and n represents the number of keywords.

[0018] Preferably, the keyword analysis result in step 2 includes the following steps:

[0019] Step 2.2: Launch the data analysis unit:

[0020] Step 2.2.1: Keyword clustering analysis: Through clustering algorithm, identify the keyword group Kp with similarity, Kp = {Kp1, Kp2, Kp j,...,Kp k}, where Kp j represents the jth keyword group, k represents the number of keyword groups, k≤n, Kp j =[Kt1,Kt2,Kt l ,...,Kt m ], Kt l represents the lth cluster keyword, m represents the number of cluster keywords, and the cluster result evaluation index Cri is obtained. b(i) is the average distance between the ith keyword and all keywords in the nearest neighbor cluster, a(i) is the average distance between the ith keyword and other keywords in the group to which it belongs, and max{} means taking the larger value of b(i) and a(i);

[0021] Step 2.2.2: Keyword clustering screening and positioning target audience: Based on the results of cluster analysis, first screen out the keyword group KP that meets the condition that the clustering result evaluation index Cri is greater than or equal to 0, KP = {KP1, KP2, KP I ,...,KP N}, KP I represents the keyword group that meets the conditions of the I-th keyword group, N represents the number of keyword groups that meet the conditions, and then locates the target audience group based on the search users of the keyword group;

[0022] Step 2.2.3: Search intent identification: Based on the keyword group KP in step 2.2.2, the keyword data in the keyword group is converted into a keyword vector through big data analysis technology, and then the cosine similarity measurement method is used to traverse the similarity between each keyword in each keyword group and the cluster center. Finally, the similarity is sorted according to the size of the similarity, and the keyword with the greatest similarity is selected as the theme of the search intent.

[0023] Preferably, the correlation analysis in step 3 includes the following steps:

[0024] Step 3.1: Based on the keyword analysis results, select keyword groups that are relevant to the target audience’s search intent. I First, through natural language processing technology, the user searches for the keyword group KP I The keywords and target digital advertising content in the image are segmented respectively; each word is then converted into a numerical vector, and the relevance between the keyword vector and the target advertising copy vector is calculated using cosine similarity traversal;

[0025] Step 3.2: Build the second keyword list: Based on the relevance analysis results, keyword group KP IThe keywords in are arranged in descending order according to the relevance, and the keywords with relevance greater than or equal to the threshold are selected to form the second keyword list Kt(2) of the target digital advertisement, Kt(2) = [Kt1, Kt2, Kt J ,...,Kt K ], where Kt J represents the keyword of the Jth target digital advertisement, and K represents the number of target digital advertisement keywords.

[0026] Preferably, step 4 comprises the following steps:

[0027] Step 4.1: Based on the second keyword list in step 3 and the search keyword data in step 1, the keyword search frequency of the target digital advertisement is obtained, and the keywords are sorted according to the search frequency to obtain the keyword group Fk of the target digital advertisement in descending order of search frequency, where Fk = [Kt1, Kt2, Kt J ,...,Kt K ], where Kt J represents the keyword of the Jth target digital advertisement after sorting, and K represents the number of keywords of the target digital advertisement;

[0028] Step 4.2: Based on the keyword search frequency of the target digital advertisement obtained in step 4.1, analyze the user browsing record data and user purchase record data corresponding to the search keyword to obtain the keyword search conversion rate Scr, Scr = Pq / Nv, where Pq represents the number of advertisement products purchased by users, and Nv represents the number of advertisement products browsed by users.

[0029] Preferably, step 4 comprises the following steps:

[0030] Step 4.3: Based on the keyword search frequency and keyword search conversion rate in step 4.1, first check the number of keyword search results returned through the search engine log, obtain the market competition level of each keyword of the target digital advertisement, and obtain the competition intensity index Cii of each keyword. Where Sf(Kt J ) represents the keyword search frequency of the Jth target digital advertisement, T_Sf(Fk) represents the total keyword search frequency, Scr(Kt J ) represents the keyword search conversion rate of the Jth target digital advertisement, Nr(Kt J ) The number of keyword search results of the Jth target digital advertisement, T_Nr(Fk) represents the total number of searches, w1, w2 and w3 represent the corresponding weights respectively; then, combined with the keywords and bidding strategies used by competitors’ advertisements, the keyword bidding dataset Kbd of the target digital advertisement is obtained according to the budget, Kbd=[Pr(Kt1),Pr(Kt2),Pr(KtJ ),...,Pr(Kt K )], where Pr(Kt J ) represents the bid value of the keyword of the Jth target digital advertisement;

[0031] Step 4.4: Sort each keyword according to its competition intensity index Cii and bid value, and select keywords whose competition intensity index is greater than or equal to the threshold and whose bid value is greater than or equal to the bid value of competitors' advertisements to form a highly competitive keyword group for the target digital advertisement. Use exact matching to deliver digital advertisements to the target user group; otherwise, use phrase matching or broad matching to deliver digital advertisements to more potential users.

[0032] Preferably, the digital advertisement delivery result in step 5 is evaluated, including the exposure Df, click count Cc, and conversion rate Scr of the digital advertisement, wherein the exposure is the number of times the advertisement is displayed, and the digital advertisement delivery effect index Aei is obtained. a1 and a2 represent the corresponding weights respectively.

[0033] Preferably, a digital advertising delivery system based on big data analysis includes:

[0034] Digital advertising data collection module: used to collect characteristic data of digital advertising, obtain various user behavior data for analyzing digital advertising methods, and transmit them to the digital advertising data processing and analysis module;

[0035] Digital advertising data processing and analysis module: used to pre-process and analyze the user behavior data in the data collection module, obtain the keyword analysis results of digital advertising, and transmit them to the digital advertising keyword optimization module;

[0036] Digital advertising keyword optimization module: used to analyze the relevance between keywords and advertising content based on the keyword analysis results in the data processing and analysis module through natural language processing technology, obtain a keyword library related to the advertising product, and generate a second keyword list at the same time, and transmit it to the digital advertising delivery module;

[0037] Digital advertisement delivery module: used for matching digital advertisement content in real time according to the second keyword list in the keyword optimization module and the user behavior data, delivering digital advertisements to the target user group, and transmitting the digital advertisement delivery results to the digital advertisement delivery effect evaluation module;

[0038] Digital advertising delivery effect evaluation module: used to evaluate the digital advertising delivery results in the digital advertising delivery module, obtain digital advertising delivery effect indicators, and transmit them to the digital advertising delivery human-computer interaction module;

[0039] Digital advertising human-computer interaction module: used to output the digital advertising effect indicators passed by the effect evaluation module to the management personnel information terminal.

[0040] Preferably, an electronic device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program in the storage device to implement a digital advertising delivery method based on big data analysis.

[0041] Technical effects and advantages of the present invention:

[0042] 1. The present invention clusters the collected advertising content keyword data through a clustering algorithm, and can identify the correlation and search intent between keywords, which helps advertisers to more accurately locate the target audience and adjust the advertising content according to the search intent to make it closer to the actual needs of users, thereby improving the click-through rate and conversion rate of advertisements and avoiding the waste of advertising resources;

[0043] 2. The present invention uses natural language processing technology to analyze the relevance of keywords and target digital advertising content, and obtains a keyword library related to the target digital advertising product, which can ensure that the advertising content is highly matched with the user's search intention, thereby helping to improve the accuracy and effect of advertising and reduce invalid clicks and conversion costs;

[0044] 3. The present invention uses big data analysis technology, combined with real-time feedback on market competition and advertising effects, and adopts a dynamic bidding strategy to automatically adjust the bidding level of keywords to obtain the maximum exposure and click-through rate of digital advertisements within a limited budget; at the same time, according to the real-time search behavior of users, the relevant keyword library is regularly updated to ensure the accuracy and effectiveness of the keyword library. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a schematic diagram of the overall process of the present invention.

[0046] Figure 2 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0047] 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.

[0048] See also Figure 1As shown, the present invention provides a digital advertising delivery method based on big data analysis, including: a digital advertising delivery data acquisition module, a digital advertising delivery data processing and analysis module, a digital advertising delivery keyword optimization module, a digital advertising delivery module, a digital advertising delivery effect evaluation module and a digital advertising delivery human-computer interaction module.

[0049] The digital advertising delivery data collection module is connected to the digital advertising delivery data processing and analysis module, the digital advertising delivery keyword optimization module is respectively connected to the digital advertising delivery data processing and analysis module and the digital advertising delivery module, the digital advertising delivery effect evaluation module is respectively connected to the digital advertising delivery module and the digital advertising delivery human-computer interaction module, and the digital advertising delivery human-computer interaction module is connected to the digital advertising delivery module.

[0050] Digital advertising data collection module: used to collect characteristic data of digital advertising, obtain various user behavior data for analyzing digital advertising methods, and transmit them to the digital advertising data processing and analysis module;

[0051] Digital advertising data processing and analysis module: used to pre-process and analyze the user behavior data in the data collection module, obtain the keyword analysis results of digital advertising, and transmit them to the digital advertising keyword optimization module;

[0052] Digital advertising keyword optimization module: used to analyze the relevance between keywords and advertising content based on the keyword analysis results in the data processing and analysis module through natural language processing technology, obtain a keyword library related to the advertising product, and generate a second keyword list at the same time, and transmit it to the digital advertising delivery module;

[0053] Digital advertisement delivery module: used for matching digital advertisement content in real time according to the second keyword list in the keyword optimization module and the user behavior data, delivering digital advertisements to the target user group, and transmitting the digital advertisement delivery results to the digital advertisement delivery effect evaluation module;

[0054] Digital advertising delivery effect evaluation module: used to evaluate the digital advertising delivery results in the digital advertising delivery module, obtain digital advertising delivery effect indicators, and transmit them to the digital advertising delivery human-computer interaction module;

[0055] Digital advertising human-computer interaction module: used to output the digital advertising effect indicators passed by the effect evaluation module to the management personnel information terminal.

[0056] See also Figure 2As shown, a digital advertising delivery method based on big data analysis includes the following steps: step 1: through log analysis and user behavior tracking technology, various user behavior data related to the advertising content are collected, and a database for analyzing the digital advertising delivery method is constructed; step 2: the various user behavior data of the digital advertising delivery method collected in step 1 are analyzed to obtain the keyword analysis results of digital advertising delivery; step 3: based on the keyword analysis results in step 2, the relevance of the keywords to the target digital advertising content is analyzed through natural language processing technology, and a keyword library related to the target digital advertising product is obtained, and a second keyword list is generated at the same time; step 4: based on the second keyword list in step 3 and the user behavior data in step 1, the digital advertising content is matched in real time, and the digital advertisement is delivered to the target user group at the same time; step 5: based on the digital advertising delivery result in step 4, the digital advertising delivery effect index is evaluated to obtain the digital advertising delivery effect index; step 6: human-computer interaction is performed on the digital advertising delivery effect index obtained in step 5.

[0057] Step 1: Through log analysis and user behavior tracking technology, collect various user behavior data related to advertising content, including user search keyword data, user browsing record data, user purchase record data, and competitor advertising data, where competitor advertising data includes the keywords used and bidding strategies, and build a database for analyzing digital advertising methods;

[0058] Step 2: Analyze various user behavior data of the digital advertising delivery method collected in step 1 to obtain keyword analysis results of digital advertising delivery, including a delivery data processing unit and a delivery data analysis unit, and the keyword analysis results include the following steps;

[0059] Step 2.1: Place the data processing unit:

[0060] Step 2.1.1: Data preprocessing: Through data preprocessing technology, the collected user behavior data is cleaned, abnormal data and invalid data are removed, thereby eliminating noise data, improving data quality, ensuring data accuracy, and obtaining data quality indicators Dqi. Where Normalization() represents the data normalization function, Ad represents abnormal data, Id represents invalid data, and Td represents total data;

[0061] It should be specifically explained in this embodiment that invalid data includes duplicate data, missing data, erroneous data, erroneous data, etc.; outliers include extreme data, outlier data, data distribution deviation data, etc. For example, the click-through rate of a certain advertisement is much higher than the average click-through rate of other advertisements; the distribution of certain features in the data set is obviously inconsistent with the actual situation, for example, the number of users in a certain area is obviously too high. These extreme values ​​may not be the actual situation and may be caused by certain errors or cheating behaviors.

[0062] Step 2.1.2: Keyword screening: Screen the user search keywords related to the advertising content from the search log. For example, if the target advertising content is "smart home", the keywords are "smart home, smart speaker, smart door lock", etc., and initially generate the first keyword list Kt(1), Kt(1) = [Kt1, Kt2, Kt i ,...,Kt n ], where Kt i represents the i-th keyword related to the target advertisement content, and n represents the number of keywords;

[0063] Step 2.2: Launch the data analysis unit:

[0064] Step 2.2.1: Keyword clustering analysis: Cluster and analyze user search keywords through clustering algorithms (such as K-means, DBSCAN, etc.), group similar keywords together, build an association network between keywords, and identify similar keyword groups Kp, Kp = {Kp1, Kp2, Kp j ,...,Kp k}, where Kp j represents the jth keyword group, k represents the number of keyword groups, k≤n, Kp j =[Kt1,Kt2,Kt l ,...,Kt m ], Kt l represents the lth cluster keyword, m represents the number of cluster keywords, for example, clustering "smart home", "smart speaker" and "smart bulb" into the "smart home product" group, and obtaining the clustering result evaluation index Cri, b(i) is the average distance between the ith keyword and all keywords in the nearest neighbor cluster, a(i) is the average distance between the ith keyword and other keywords in the group to which it belongs, max{} means taking the larger value of b(i) and a(i), and the value range of the clustering result evaluation index Cri is [-1,1]. The closer the value is to 1, the better the clustering result is, and the closer the value is to -1, the worse the clustering result is;

[0065] Step 2.2.2: Keyword clustering screening and positioning target audience: Based on the results of cluster analysis, first screen out the keyword group KP that meets the condition that the clustering result evaluation index Cri is greater than or equal to 0, KP = {KP1, KP2, KP I ,...,KP N}, KP I represents the keyword group that meets the conditions of the I-th keyword group, N represents the number of keyword groups that meet the conditions, and then locates the target audience group based on the search users of the keyword group;

[0066] Step 2.2.3: Search intent identification: Based on the keyword group KP in step 2.2.2, the keyword data in the keyword group is converted into a keyword vector through big data analysis technology. Then, the cosine similarity measurement method is used to traverse the similarity between each keyword in each keyword group and the cluster center. Finally, the similarity is sorted according to the size, and the keyword with the greatest similarity is selected as the theme of the search intent.

[0067] What needs to be specifically explained in this embodiment is that through cluster analysis of keywords, user groups with similar search intents or interest preferences can be identified, which helps advertisers to more accurately locate target audiences and understand users' search intents. Advertisers can adjust advertising content based on these intents to make it more in line with users' actual needs, thereby improving the click-through rate and conversion rate of advertisements and avoiding waste of advertising resources.

[0068] Step 3: Based on the keyword analysis results in step 2, the relevance between the keywords and the target digital advertisement content is analyzed by natural language processing technology to obtain a keyword library related to the target digital advertisement product and generate a second keyword list. The relevance analysis includes the following steps:

[0069] Step 3.1: Based on the keyword analysis results, select keyword groups that are relevant to the target audience’s search intent. I First, through natural language processing technology, the user searches for the keyword group KP I The keywords and target digital advertising content in the article are segmented separately. For example, the keyword "smart home security" is segmented into "smart", "home", and "security", and the target digital advertising content "smart door lock, protect your home" is segmented into "smart", "door lock", "protect", "your", and "home". Each word is then converted into a numerical vector, and the cosine similarity is used to traverse and calculate the correlation between the keyword vector and the vector of the target advertising copy. The closer the value is to 1, the higher the correlation, and vice versa. Using natural language processing technology to analyze the correlation between keywords and advertising content can help us match keywords and advertising content more accurately, optimize advertising copy, improve advertising effects, and ensure that advertising can accurately reach potential users.

[0070] Step 3.2: Build the second keyword list: Based on the relevance analysis results, keyword group KP I The keywords in are arranged in descending order according to the relevance, and the keywords with relevance greater than or equal to the threshold are selected to form the second keyword list Kt(2) of the target digital advertisement, Kt(2) = [Kt1, Kt2, Kt J ,...,Kt K ], where Kt J represents the keyword of the Jth target digital advertisement, K represents the number of the target digital advertisement keywords, for example, keywords with a relevance greater than 0.7 belong to the second keyword list of the target digital advertisement;

[0071] What needs to be specifically explained in this embodiment is that, as the advertisement content is updated, the relevant keyword library is updated regularly to ensure the accuracy and effectiveness of the keyword library.

[0072] Step 4: Based on the second keyword list in step 3 and the user behavior data in step 1, including user search keyword data, user browsing record data, user purchase record data, and competitor advertising data, where the competitor advertising data includes the keywords used and bidding strategies, real-time matching of digital advertising content is performed, and digital advertising is delivered to the target user group. The digital advertising delivery includes the following steps:

[0073] Step 4.1: Based on the second keyword list in step 3 and the search keyword data in step 1, the keyword search frequency of the target digital advertisement is obtained, and the keywords are sorted according to the search frequency to obtain the keyword group Fk of the target digital advertisement in descending order of search frequency, where Fk = [Kt1, Kt2, Kt J ,...,Kt K ], where Kt J represents the keyword of the Jth target digital advertisement after sorting, and K represents the number of keywords of the target digital advertisement;

[0074] Step 4.2: Based on the keyword search frequency of the target digital advertisement obtained in step 4.1, analyze the user browsing record data and user purchase record data corresponding to the search keyword, the user purchase record data includes the data of purchase, registration, download, etc., and obtain the keyword search conversion rate Scr, Scr = Pq / Nv, where Pq represents the number of advertisement products purchased by users, and Nv represents the number of advertisement products browsed by users;

[0075] Step 4.3: Based on the keyword search frequency and keyword search conversion rate in step 4.1, determine the keyword bidding strategy. First, check the number of keyword search results returned through the search engine log, obtain the market competition level of each keyword of the target digital advertisement, and obtain the competition intensity index Cii of each keyword. Where Sf(Kt J ) represents the keyword search frequency of the Jth target digital advertisement, T_Sf(Fk) represents the total keyword search frequency, Scr(Kt J ) represents the keyword search conversion rate of the Jth target digital advertisement, Nr(Kt J ) The number of keyword search results of the Jth target digital advertisement, T_Nr(Fk) represents the total number of searches, w1, w2 and w3 represent the corresponding weights, for example, w1=0.4, w2=0.3 and w3=0.3; then, combined with the keywords and bidding strategies used by competitors’ advertisements, the keyword bidding dataset Kbd of the target digital advertisement is obtained according to the budget, Kbd=[Pr(Kt1), Pr(Kt2), Pr(Kt J ),...,Pr(Kt K )], where Pr(Kt J ) represents the bid value of the keyword of the Jth target digital advertisement;

[0076] What needs to be specifically explained in this embodiment is that the higher the keyword bid, the more opportunities the advertisement will generally have to be displayed on search engines or other advertising platforms, and the higher the advertisement will generally be ranked in the search results, which helps to attract more users' attention and clicks, and can help advertisers stand out in keyword competition, suppress competitors' ad displays, and thus seize more market share and user attention.

[0077] Step 4.4: Sort each keyword according to its competitive intensity index Cii and bid value, select keywords whose competitive intensity index is greater than or equal to the threshold and whose bid value is greater than or equal to the bid value of competitor's advertisement, form a competitive keyword group of target digital advertisement, and deliver digital advertisement to target user group by using exact matching method. Exact matching means that when the keyword searched by the user is exactly the same as the keyword of the target digital advertisement, the advertisement will be displayed to the relevant user group; otherwise, phrase matching or broad matching method is used to deliver digital advertisement to more potential users. Phrase matching means that when the keyword searched by the user contains the keyword of the target digital advertisement, the advertisement may be displayed, and broad matching means that when the keyword searched by the user is related to the keyword of the target digital advertisement, the advertisement may be displayed;

[0078] Step 5: Based on the results of digital advertising in step 4, the advertising effect is evaluated by using data monitoring tools, including the exposure volume Df, click count Cc, and conversion rate Scr of the digital advertisement. The exposure volume is the number of times the advertisement is displayed, and the digital advertising effect index Aei is obtained. a1 and a2 represent the corresponding weights, for example, a1 = 0.5, a2 = 0.5;

[0079] Step 6: Conduct human-computer interaction on the digital advertising delivery effect indicators obtained in step 5. When the digital advertising delivery effect indicators are within the set allowable range, it means that the digital advertising delivery method is effective. Otherwise, it prompts the management personnel to take timely measures to iteratively optimize the advertising delivery strategy to form a closed-loop feedback mechanism. Combined with the market competition situation and real-time feedback on advertising effects, a dynamic bidding strategy is adopted to automatically adjust the bidding level of keywords to obtain maximum exposure and click-through rate within a limited budget. For example, A / B tests are run regularly to test the effects of different keywords and advertising content, and the delivery strategy is optimized through comparative analysis.

[0080] It should be specifically explained that the present invention also provides an electronic device. In an electronic device embodiment according to the present invention, the electronic device includes a processor and a storage device. The storage device can be configured to store a computer program for executing the digital advertising delivery method of the above method embodiment. The processor can be configured to execute the program in the storage device, which includes but is not limited to the program for executing the digital advertising delivery method of the above method embodiment. The electronic device can be a control device device formed by various electronic devices.

[0081] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0082] The above contents are merely examples and explanations of the concept of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. A digital advertising delivery method based on big data analysis, characterized in that: The following steps are involved: Step 1: Collect various user behavior data related to advertising content through log analysis and user behavior tracking technology, and build a database for analyzing digital advertising delivery methods; Step 2: Analyze various user behavior data of the digital advertising delivery method collected in step 1 to obtain keyword analysis results of digital advertising delivery; Step 3: Based on the keyword analysis results in step 2, the relevance between the keywords and the target digital advertisement content is analyzed by natural language processing technology to obtain a keyword library related to the target digital advertisement product, and generate a second keyword list; Step 4: Based on the second keyword list in step 3 and the user behavior data in step 1, digital advertisement content is matched in real time, and digital advertisements are delivered to target user groups; Step 5: Evaluate the digital advertising delivery results in step 4 to obtain digital advertising delivery effect indicators; Step 6: Conduct human-computer interaction on the digital advertising effect indicators obtained in step 5.

2. A digital advertising delivery method based on big data analysis according to claim 1, characterized in that: The user behavior data in step 2 include user search keyword data, user browsing record data, user purchase record data, and competitor advertising data, wherein the competitor advertising data includes the keywords used and the bidding strategy.

3. The digital advertising delivery method based on big data analysis according to claim 1, characterized in that: The keyword analysis result in step 2 includes the following steps: Step 2.1: Place the data processing unit: Step 2.1.1: Data preprocessing: Through data preprocessing technology, the data quality index Dqi is obtained. Where Normalization() represents the data normalization function, Ad represents abnormal data, Id represents invalid data, and Td represents total data; Step 2.1.2: Keyword screening: Screen the user search keywords related to the advertising content from the search log, and preliminarily generate the first keyword list Kt(1), Kt(1) = [Kt1, Kt2, Kt i ,...,Kt n ], where Kt i represents the ith keyword related to the target advertisement content, and n represents the number of keywords.

4. The digital advertising delivery method based on big data analysis according to claim 1, characterized in that: The keyword analysis result in step 2 includes the following steps: Step 2.2: Launch the data analysis unit: Step 2.2.1: Keyword clustering analysis: Through clustering algorithm, identify the keyword group Kp with similarity, Kp = {Kp1, Kp2, Kp j ,...,Kp k }, where Kp j represents the jth keyword group, k represents the number of keyword groups, k≤n, Kp j =[Kt1,Kt2,Kt l ,...,Kt m ], Kt l represents the lth cluster keyword, m represents the number of cluster keywords, and the cluster result evaluation index Cri is obtained. b(i) is the average distance between the ith keyword and all keywords in the nearest neighbor cluster, a(i) is the average distance between the ith keyword and other keywords in the group to which it belongs, and max{} means taking the larger value of b(i) and a(i); Step 2.2.2: Keyword clustering screening and positioning target audience: Based on the results of cluster analysis, first screen out the keyword group KP that meets the condition that the clustering result evaluation index Cri is greater than or equal to 0, KP = {KP1, KP2, KP I ,...,KP N }, KP I represents the keyword group that meets the conditions of the I-th keyword group, N represents the number of keyword groups that meet the conditions, and then locates the target audience group based on the search users of the keyword group; Step 2.2.3: Search intent identification: Based on the keyword group KP in step 2.2.2, the keyword data in the keyword group is converted into a keyword vector through big data analysis technology, and then the cosine similarity measurement method is used to traverse the similarity between each keyword in each keyword group and the cluster center. Finally, the similarity is sorted according to the size of the similarity, and the keyword with the greatest similarity is selected as the theme of the search intent.

5. The digital advertising delivery method based on big data analysis according to claim 1, characterized in that: The correlation analysis in step 3 includes the following steps: Step 3.1: Based on the keyword analysis results, select keyword groups that are relevant to the target audience’s search intent. I First, through natural language processing technology, the user searches for the keyword group KP I The keywords and target digital advertising content in the image are segmented respectively; each word is then converted into a numerical vector, and the relevance between the keyword vector and the target advertising copy vector is calculated using cosine similarity traversal; Step 3.2: Build the second keyword list: Based on the relevance analysis results, keyword group KP I The keywords in are arranged in descending order according to the relevance, and the keywords with relevance greater than or equal to the threshold are selected to form the second keyword list Kt(2) of the target digital advertisement, Kt(2) = [Kt1, Kt2, Kt J ,...,Kt K ], where Kt J represents the keyword of the Jth target digital advertisement, and K represents the number of target digital advertisement keywords.

6. The digital advertising delivery method based on big data analysis according to claim 1, characterized in that: The step 4 comprises the following steps: Step 4.1: Based on the second keyword list in step 3 and the search keyword data in step 1, the keyword search frequency of the target digital advertisement is obtained, and the keywords are sorted according to the search frequency to obtain the keyword group Fk of the target digital advertisement in descending order of search frequency, where Fk = [Kt1, Kt2, Kt J ,...,Kt K ], where Kt J represents the keyword of the Jth target digital advertisement after sorting, and K represents the number of keywords of the target digital advertisement; Step 4.2: Based on the keyword search frequency of the target digital advertisement obtained in step 4.1, analyze the user browsing record data and user purchase record data corresponding to the search keyword to obtain the keyword search conversion rate Scr, Scr = Pq / Nv, where Pq represents the number of advertisement products purchased by users, and Nv represents the number of advertisement products browsed by users.

7. The digital advertising delivery method based on big data analysis according to claim 1, characterized in that: The step 4 comprises the following steps: Step 4.3: Based on the keyword search frequency and keyword search conversion rate in step 4.1, first check the number of keyword search results returned through the search engine log, obtain the market competition level of each keyword of the target digital advertisement, and obtain the competition intensity index Cii of each keyword. Where Sf(Kt J ) represents the keyword search frequency of the Jth target digital advertisement, T_Sf(Fk) represents the total keyword search frequency, Scr(Kt J ) represents the keyword search conversion rate of the Jth target digital advertisement, Nr(Kt J ) The number of keyword search results of the Jth target digital advertisement, T_Nr(Fk) represents the total number of searches, w1, w2 and w3 represent the corresponding weights respectively; then, combined with the keywords and bidding strategies used by competitors’ advertisements, the keyword bidding dataset Kbd of the target digital advertisement is obtained according to the budget, Kbd=[Pr(Kt1),Pr(Kt2),Pr(Kt J ),...,Pr(Kt K )], where Pr(Kt J ) represents the bid value of the keyword of the Jth target digital advertisement; Step 4.4: Sort each keyword according to its competition intensity index Cii and bid value, and select keywords whose competition intensity index is greater than or equal to the threshold and whose bid value is greater than or equal to the bid value of competitors' advertisements to form a highly competitive keyword group for the target digital advertisement. Use exact matching to deliver digital advertisements to the target user group; otherwise, use phrase matching or broad matching to deliver digital advertisements to more potential users.

8. The digital advertising delivery method based on big data analysis according to claim 1, characterized in that: In step 5, the digital advertisement delivery result is evaluated, including the exposure amount Df, the number of clicks Cc, and the conversion rate Scr of the digital advertisement. The exposure amount is the number of times the advertisement is displayed, and the digital advertisement delivery effect index Aei is obtained. a1 and a2 represent the corresponding weights respectively.

9. A digital advertising delivery system based on big data analysis, used to use the digital advertising delivery method based on big data analysis described in any one of claims 1 to 8, comprising: Digital advertising data collection module: used to collect characteristic data of digital advertising, obtain various user behavior data for analyzing digital advertising methods, and transmit them to the digital advertising data processing and analysis module; Digital advertising data processing and analysis module: used to pre-process and analyze the user behavior data in the data collection module, obtain the keyword analysis results of digital advertising, and transmit them to the digital advertising keyword optimization module; Digital advertising keyword optimization module: used to analyze the relevance between keywords and advertising content based on the keyword analysis results in the data processing and analysis module through natural language processing technology, obtain a keyword library related to the advertising product, and generate a second keyword list at the same time, and transmit it to the digital advertising delivery module; Digital advertisement delivery module: used for matching digital advertisement content in real time according to the second keyword list in the keyword optimization module and the user behavior data, delivering digital advertisements to the target user group, and transmitting the digital advertisement delivery results to the digital advertisement delivery effect evaluation module; Digital advertising delivery effect evaluation module: used to evaluate the digital advertising delivery results in the digital advertising delivery module, obtain digital advertising delivery effect indicators, and transmit them to the digital advertising delivery human-computer interaction module; Digital advertising human-computer interaction module: used to output the digital advertising effect indicators passed by the effect evaluation module to the management personnel information terminal.

10. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program in the storage device to implement a digital advertising delivery method based on big data analysis as described in any one of claims 1 to 8.

Images