CN118071427A - Advertisement preview analysis method, system, processor and storage medium - Google Patents
Advertisement preview analysis method, system, processor and storage medium Download PDFInfo
- Publication number
- CN118071427A CN118071427A CN202410300980.0A CN202410300980A CN118071427A CN 118071427 A CN118071427 A CN 118071427A CN 202410300980 A CN202410300980 A CN 202410300980A CN 118071427 A CN118071427 A CN 118071427A
- Authority
- CN
- China
- Prior art keywords
- advertisement
- user
- preview
- prediction model
- preliminary
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 58
- 238000003860 storage Methods 0.000 title claims abstract description 9
- 230000000694 effects Effects 0.000 claims abstract description 68
- 238000000034 method Methods 0.000 claims abstract description 48
- 238000011156 evaluation Methods 0.000 claims abstract description 37
- 238000005457 optimization Methods 0.000 claims abstract description 29
- 230000006399 behavior Effects 0.000 claims description 100
- 230000007246 mechanism Effects 0.000 claims description 12
- 230000003993 interaction Effects 0.000 claims description 11
- 230000004044 response Effects 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 238000004140 cleaning Methods 0.000 claims description 7
- 238000012544 monitoring process Methods 0.000 claims description 7
- 230000003213 activating effect Effects 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 6
- 238000013480 data collection Methods 0.000 claims description 4
- 238000007418 data mining Methods 0.000 claims description 4
- 238000010801 machine learning Methods 0.000 claims description 4
- 238000010206 sensitivity analysis Methods 0.000 claims description 4
- 230000002452 interceptive effect Effects 0.000 claims description 3
- 238000007726 management method Methods 0.000 claims description 3
- 230000004043 responsiveness Effects 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 6
- 238000012384 transportation and delivery Methods 0.000 description 5
- 238000012795 verification Methods 0.000 description 3
- 230000002996 emotional effect Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000003631 expected effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
- G06Q30/0271—Personalized advertisement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0255—Targeted advertisements based on user history
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method, a system, a processor and a storage medium for advertisement preview analysis, which comprise the steps of subdividing advertisement contents according to the characteristics of the pre-placed advertisement contents to generate advertisement feature labels, and generating user behavior feature labels through a user behavior analysis module; evaluating the matching degree of each user behavior feature tag and each advertisement feature tag through a preset matching algorithm; according to the matching degree evaluation result, invoking an advertisement effect prediction model to provide a preliminary advertisement preview strategy; when a user interacts with advertisement previews, collecting feedback information of the user in real time through a data collecting unit; continuously optimizing and updating the advertisement effect prediction model; the preliminary advertisement preview strategy is optimized based on the optimized and updated advertisement effect prediction model to generate corresponding advertisement putting strategy suggestions, and the continuous optimization flow enables advertisement content to remarkably improve user satisfaction and brings higher user responsiveness to advertisements.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a method, a system, a processor, and a storage medium for advertisement preview analysis.
Background
With the continuous development of big data and artificial intelligence technology, the advertising industry is undergoing a profound revolution, advertisers and marketers are looking for more advanced methods to increase the conversion rate and the return rate of advertisements; furthermore, the importance of user experience is increasing, and the personalized demand for advertising content and the accurate prediction of preview effects become an integral part of marketing strategies.
The existing advertisement preview technology mainly focuses on simple user grouping based on user history behavior data and then displaying related advertisement previews; although the method improves the pertinence of the advertisement to a certain extent, obvious defects still exist: one is that the user grouping is too static and lacks real-time response to the evolution of the user interests; secondly, the generation and optimization of advertisement previews lack individuation, and the specific preview scene and equipment characteristics of users cannot be fully considered, so that fluctuation and uncertainty of advertisement effects are caused; these prior art drawbacks have resulted in a far from optimal advertising effectiveness and effectiveness.
In view of this, there is a need for improvements in the art of advertisement preview technology to address the technical shortcomings of advertisement delivery lacking personalized designs.
Disclosure of Invention
The invention aims to provide a method, a system, a processor and a storage medium for analyzing advertisement previews, which solve the technical problems.
To achieve the purpose, the invention adopts the following technical scheme:
A method of advertisement preview analysis, comprising:
according to the characteristics of the pre-placed advertisement content, subdividing the advertisement content to generate advertisement feature labels;
activating a user behavior analysis module by analyzing historical browsing records, application using habits and feedback data of a user, and generating a user behavior feature tag through the user behavior analysis module;
Evaluating the matching degree of each user behavior feature tag and each advertisement feature tag through a preset matching algorithm;
according to the matching degree evaluation result, invoking an advertisement effect prediction model to provide a preliminary advertisement preview strategy;
when a user interacts with advertisement previews, collecting feedback information of the user in real time through a data collecting unit;
continuously optimizing and updating the advertisement effect prediction model by utilizing feedback data collected in real time;
And optimizing the preliminary advertisement preview strategy based on the optimized and updated advertisement effect prediction model to generate corresponding advertisement putting strategy suggestions.
Optionally, the advertisement content is subdivided according to the characteristics of the pre-delivered advertisement content, and an advertisement feature tag is generated; the method specifically comprises the following steps:
classifying the pre-placed advertisement content according to the main types of the advertisement content, and extracting and analyzing key attributes of each type of advertisement content;
Based on the key attributes of the advertisements, analyzing target groups, and analyzing to obtain user portraits expected to be touched by the advertisements;
associating the collected key attributes of the advertisement and the target group information with a specific consumption context;
and generating a preliminary advertisement feature label according to the classification of the advertisement content, the key attribute, the user portrait and the context associated information.
Optionally, the advertisement content is subdivided according to the characteristics of the pre-delivered advertisement content, and an advertisement feature tag is generated; further comprises:
Verifying the generated preliminary advertisement feature labels in a small-scale target user group, and collecting preliminary feedback information;
and verifying and optimizing the preliminary advertisement feature labels through the preliminary feedback information to generate advertisement feature labels, and integrating the advertisement feature labels into a continuously updated label library.
Optionally, the user behavior analysis module is activated by analyzing the historical browsing record, the application using habit and the feedback data of the user, and the user behavior feature tag is generated by the user behavior analysis module; the method specifically comprises the following steps:
collecting historical browsing records, application using habits and feedback data of historical advertisement contents of a user in a concentrated manner to obtain a first data packet;
performing preliminary cleaning processing on the first data packet, wherein the cleaning processing comprises the step of removing invalid or wrong data records;
identifying a main behavior pattern of a user from the cleaned first data packet by using a deep machine learning data mining technology;
deeply analyzing user feedback data of the user on the advertisement content to analyze preference and response intensity of the user on different types of advertisement content;
Activating a user behavior analysis module, and generating a group of user behavior feature labels for each user based on the identified main behavior mode and user feedback data;
Grouping users according to the generated behavior feature labels, and periodically updating the behavior feature labels according to the latest behavior mode of the users and user feedback data by using a real-time feedback circulation mechanism;
And integrating the behavior feature labels of all the users into a dynamically updated behavior feature database.
Optionally, the matching degree between each user behavior feature tag and each advertisement feature tag is evaluated through a preset matching algorithm; the method specifically comprises the following steps:
Selecting a preset matching algorithm, and setting weights of the user behavior feature tags and the advertisement feature tags;
Calculating the matching degree of each user behavior characteristic label and each advertisement characteristic label by using a selected matching algorithm;
analyzing the preliminary result of the matching degree evaluation operation, and identifying the user with the highest score to be paired with the advertisement;
The weights of the user behavior feature tags and the advertisement feature tags are adjusted to conduct sensitivity analysis on a matching algorithm, so that the influence degree of tag weight change on matching degree scoring is evaluated, and the matching algorithm is optimized;
After weight adjustment and optimization of a matching algorithm, carrying out matching degree evaluation operation again, and confirming a final matching degree score;
And recording the final matching degree evaluation result, and transmitting the final matching degree evaluation result to the advertisement effect prediction model.
Optionally, according to the matching degree evaluation result, invoking an advertisement effect prediction model to provide a preliminary advertisement preview strategy; the method specifically comprises the following steps:
According to the matching degree evaluation result, an advertisement effect prediction model is called, and a preliminary advertisement preview strategy framework is constructed through the advertisement effect prediction model;
adding user experience indexes into a preliminary advertisement preview policy framework to generate a preliminary advertisement preview policy;
Setting an adjustment mechanism in the preliminary advertisement preview strategy, predefining an adjustment parameter, and starting the adjustment mechanism to adjust the preliminary advertisement preview strategy when user feedback data triggers the adjustment parameter.
Optionally, continuously optimizing and updating the advertisement effect prediction model by using feedback data collected in real time; the method specifically comprises the following steps:
tracking and collecting feedback data of advertisement preview interaction of a user through a data collecting unit;
Summarizing and sorting feedback data collected in real time, and taking the feedback data as an input variable of an optimized advertisement effect prediction model;
based on feedback data collected in real time, evaluating the current performance index of the advertisement effect prediction model, and formulating an optimization scheme according to the evaluation result of the performance index;
and inputting feedback data collected in real time into the advertisement effect prediction model, and retraining the advertisement effect prediction model according to an optimization scheme and the feedback data so as to continuously optimize and update the advertisement effect prediction model.
The invention provides an advertisement preview analysis system, which is used for realizing the method of advertisement preview analysis; the advertisement preview analysis system specifically comprises:
the advertisement feature management unit is used for subdividing advertisement contents to generate advertisement feature labels;
The data collection unit is used for collecting historical browsing records, application using habits and feedback data of the user;
The user behavior analysis module is used for generating a user behavior characteristic tag according to the historical browsing record, the application using habit and the feedback data of the user;
The data processing unit is used for storing a preset matching algorithm and an advertisement effect prediction model; the preset matching algorithm is used for evaluating the matching degree of each user behavior characteristic label and each advertisement characteristic label, and the advertisement effect prediction model is used for generating a preliminary advertisement preview strategy according to the matching degree evaluation result;
the strategy optimization unit is used for continuously optimizing and updating the advertisement effect prediction model according to feedback data collected in real time;
and the monitoring unit comprises an interactive interface and is used for monitoring and presenting the running state of the advertisement preview analysis system.
The invention provides a processor, which comprises a memory and at least one processor, wherein the memory stores instructions;
the processor invokes the instructions in the memory to cause the processor to perform the method of advertisement preview analysis as described above.
The present invention provides a storage medium having stored thereon instructions for implementing a method of advertisement preview analysis as described above.
Compared with the prior art, the invention has the following beneficial effects: during work, the pre-put advertisement content is subdivided to generate detailed advertisement feature labels, and a user behavior analysis module is activated to generate user behavior feature labels by analyzing historical browsing records, application using habits and feedback data of users; a preset matching algorithm is adopted to carry out fine evaluation on the matching degree between the user characteristic tag and the advertisement characteristic tag; based on the matching degree evaluation result, invoking an advertisement effect prediction model to provide a preliminary advertisement preview strategy; feedback data generated by interaction of a user and the preview is collected by the system immediately, so that real-time data is provided for the model, and self-optimization and updating can be performed; finally, the periodical optimized advertisement effect prediction model adjusts the preliminary advertisement preview strategy to generate a final advertisement putting strategy suggestion; the method ensures the high matching between the advertisement and the user through detailed user analysis and advertisement content subdivision, thereby obviously improving the individuation level of the advertisement and the acceptance of the user; the collection of real-time feedback and model optimization ensure that the dynamic adjustment of the advertisement strategy can quickly respond to the changes of markets and user behaviors, thereby improving the efficiency and effect of advertisement delivery; in addition, the continuous optimization flow enables advertisement content to be developed synchronously with user behaviors and favorites, so that user satisfaction can be remarkably improved in long-term view, and higher user responsiveness is brought to advertisements.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained from these drawings without inventive faculty for a person skilled in the art.
The structures, proportions, sizes, etc. shown in the drawings are shown only in connection with the present disclosure, and are not intended to limit the scope of the invention, since any modification, variation in proportions, or adjustment of the size, etc. of the structures, proportions, etc. should be considered as falling within the spirit and scope of the invention, without affecting the effect or achievement of the objective.
FIG. 1 is a flow chart of a method for analyzing advertisement previews according to the first embodiment;
FIG. 2 is a second flow chart of a method for analyzing advertisement previews according to the first embodiment;
fig. 3 is a third flowchart illustrating a method of advertisement preview analysis according to the first embodiment.
Detailed Description
In order to make the objects, features and advantages of the present invention more comprehensible, the technical solutions in the embodiments of the present invention are described in detail below with reference to the accompanying drawings, and it is apparent that the embodiments described below are only some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, it should be understood that the directions or positional relationships indicated by the terms "upper", "lower", "top", "bottom", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention. It is noted that when one component is referred to as being "connected" to another component, it can be directly connected to the other component or intervening components may also be present.
The technical scheme of the invention is further described below by the specific embodiments with reference to the accompanying drawings.
Embodiment one:
referring to fig. 1 to 3, an embodiment of the present invention provides a method for analyzing advertisement previews, including:
s1, subdividing advertisement content according to the characteristics of the pre-cast advertisement content to generate advertisement feature labels; these advertising feature tags include not only attributes of the product or service, but also expected user response and emotional tendencies;
This step focuses on understanding and subdividing the advertising content, not only to take into account the physical properties of the product or service, but also to include the intended user response and emotional tendency. The multi-dimensional tagging ensures that the advertisement preview system can accurately understand the uniqueness of advertisement content and resonance that the advertisement content may generate with a specific user group; this preliminary step is the basis for building an effective advertising strategy.
S2, activating a user behavior analysis module by analyzing historical browsing records, application using habits and feedback data of a user, and generating a user behavior feature tag through the user behavior analysis module; these user behavior feature tags reflect the user's interest preferences, active time periods, and interaction preferences.
By analyzing the user's advertisement browsing behavior, including browsing history, application usage habits, and previous feedback, the system generates tags regarding user interests, campaign periods, and interaction preferences; this allows the system to accurately outline the behavior of each user, providing the basis for subsequent personalized advertisement previews.
S3, evaluating the matching degree of each user behavior feature tag and each advertisement feature tag through a preset matching algorithm;
In this step, a preset matching algorithm is used to evaluate the degree of matching between the user behavior feature tag and the advertisement feature tag; such evaluation aims at determining which advertising content is most likely to be of interest to a particular user population, thereby ensuring relevance and efficiency of advertising.
S4, according to the matching degree evaluation result, invoking an advertisement effect prediction model to provide a preliminary advertisement preview strategy;
Based on the result of the matching degree evaluation, the advertisement effect prediction model provides a preliminary advertisement preview strategy; this strategy takes into account the expected effectiveness of the advertisement and attempts to recommend advertisement content to the target user that is most appropriate, i.e., most likely to produce a positive response.
S5, when the user interacts with the advertisement preview, collecting feedback information of the user in real time through a data collecting unit; the feedback information comprises click rate, watching duration and data of interaction behavior;
When a user interacts with advertisement content, the system collects data about click rate, watching duration and interaction behavior in real time; these real-time feedback information are an indispensable data source for optimizing advertisement strategies and predictive models.
S6, continuously optimizing and updating the advertisement effect prediction model by utilizing feedback data collected in real time;
Using the collected real-time feedback data, the system constantly optimizes and updates the advertisement effectiveness prediction model, which ensures that the model can learn and adapt to changes in user behavior, maintaining the accuracy and efficiency of its predictive capabilities.
And S7, optimizing the preliminary advertisement preview strategy based on the optimized and updated advertisement effect prediction model so as to generate a corresponding advertisement putting strategy suggestion.
Finally, according to the optimized and updated advertisement effect prediction model, the system further optimizes a preliminary advertisement preview strategy to generate a more accurate advertisement putting strategy suggestion; this ensures that the advertising strategy is continually adapted to changes in user behavior over time, optimizing the effectiveness of the advertising.
The working principle of the invention is as follows: during work, the pre-put advertisement content is subdivided to generate detailed advertisement feature labels, and a user behavior analysis module is activated to generate user behavior feature labels by analyzing historical browsing records, application using habits and feedback data of users; a preset matching algorithm is adopted to carry out fine evaluation on the matching degree between the user characteristic tag and the advertisement characteristic tag; based on the matching degree evaluation result, invoking an advertisement effect prediction model to provide a preliminary advertisement preview strategy; feedback data generated by interaction of a user and the preview is collected by the system immediately, so that real-time data is provided for the model, and self-optimization and updating can be performed; finally, the periodical optimized advertisement effect prediction model adjusts the preliminary advertisement preview strategy to generate a final advertisement putting strategy suggestion; compared with the advertisement previewing technology in the prior art, the method ensures the high matching between the advertisement and the user through detailed user analysis and advertisement content subdivision, thereby remarkably improving the individuation level of the advertisement and the acceptance of the user; the collection of real-time feedback and model optimization ensure that the dynamic adjustment of the advertisement strategy can quickly respond to the changes of markets and user behaviors, thereby improving the efficiency and effect of advertisement delivery; in addition, the continuous optimization flow enables advertisement content to be developed synchronously with user behaviors and favorites, so that user satisfaction can be remarkably improved in long-term view, and higher user responsiveness is brought to advertisements.
In this embodiment, specifically, referring to fig. 2, step S1 specifically includes:
S11, classifying the pre-placed advertisement content according to the main types of the advertisement content, and extracting and analyzing key attributes of each type of advertisement content;
Classifying advertisement content according to main types, such as video advertisement, picture advertisement and text advertisement; for each type of advertisement content, further extracting and analyzing key attributes of the advertisement content, wherein the key attributes comprise visual elements, text content and calling actions of the advertisement; this is a preliminary step in identifying and understanding the core features of the advertising content.
S12, analyzing a target group based on key attributes of the advertisement, and analyzing to obtain a user portrait expected to be touched by the advertisement;
After understanding the key attributes of the advertisement, this step aims at an analysis of the target user population (i.e., user portraits) based on the advertisement attributes, including analysis of the characteristics of the age, gender, hobbies of interest, purchasing power of the user that the advertisement content is expected to reach, which helps to clarify the location and target market of the advertisement.
S13, associating the collected key attributes and target group information of the advertisements with specific consumption situations;
Associating the collected advertisement key attributes and target user group information with a specific consumption context; this step takes into account the specific context and context of the advertising user, such as time, place, device used. Contextual relevance is a key step in improving advertising appeal and effectiveness, as it helps ensure that the advertising content is highly relevant to the user's current context.
S14, generating a preliminary advertisement feature label according to the classification of advertisement content, the key attribute, the user portrait and the context associated information.
Combining the previous analysis, and generating a preliminary advertisement feature tag set according to the classification of advertisement content, key attributes, target user portraits and context association; these ad feature tags will be used to describe the overall characteristics of the ad, which are critical to subsequent ad matching analysis and ad effectiveness prediction.
S15, verifying the generated preliminary advertisement feature labels in a small-scale target user group, and collecting preliminary feedback information;
Carrying out actual verification on the generated preliminary advertisement feature labels in a small-scale target user group, wherein the process comprises the steps of testing the attraction of advertisements, the interaction behavior of users and feedback collection; feedback information during the verification stage is critical to confirm the accuracy and validity of the advertising label.
S16, verifying and optimizing the preliminary advertisement feature labels through the preliminary feedback information to generate advertisement feature labels, and integrating the advertisement feature labels into a continuously updated label library.
And (3) carrying out refinement verification and optimization on the advertisement feature labels generated preliminarily based on the preliminary feedback information collected in the step (S15). The optimized advertisement feature labels more accurately reflect the association degree of advertisement content and target users. Finally, integrating the optimized advertisement feature labels into a label library capable of being updated continuously so as to facilitate continuous use and reference of the advertisement preview analysis system
In this embodiment, specifically, referring to fig. 3, step S2 specifically includes:
S21, collecting historical browsing records, application using habits and feedback data of historical advertisement contents of a user in a concentrated manner to obtain a first data packet;
This step involves the centralized collection of extensive data about the user, including but not limited to historical browsing records, application usage habits, and user feedback information on historical advertising content. These data form the first data package and are the basis for analyzing the user behavior and optimizing the advertising strategy.
S22, performing preliminary cleaning treatment on the first data packet, wherein the cleaning treatment comprises invalid or wrong data record elimination; to eliminate abnormal browsing behavior such as accidental clicking misoperation; the data quality of subsequent analysis processing is ensured, and the analysis accuracy is improved.
The collected data often contains some noise, such as erroneous records, repeated information, or irrelevant data, and the cleaning of the first data packet, the removal of invalid or erroneous records, is a necessary step to ensure data quality and analysis accuracy.
S23, recognizing a main behavior mode of a user from the cleaned first data packet by using a data mining technology of deep machine learning;
This step uses deep machine learning and data mining techniques to identify the user's primary behavior patterns from the cleaned data, including key behavior features of the user's browsing preferences, activity time, frequency, providing a high quality input for generating user behavior tags.
S24, deeply analyzing user feedback data of the user on the advertisement content to analyze preference and response intensity of the user on different types of advertisement content;
By deeply analyzing the feedback of the user on the advertisement content, the preference and the response strength of the user on different types of advertisement content can be abstracted, the specific requirements and feelings of the user can be understood, and the basis is provided for personalized advertisement recommendation.
S25, activating a user behavior analysis module, and generating a group of user behavior feature labels for each user based on the identified main behavior mode and user feedback data;
Based on the identified user behavior patterns and the user feedback data, the user behavior analysis module is activated to generate a set of detailed user behavior feature tags for each user.
S26, grouping users according to the generated behavior feature labels, and periodically updating the behavior feature labels according to the latest behavior mode of the users and user feedback data by using a real-time feedback circulation mechanism;
Dividing users into different groups according to the generated behavior feature labels so as to realize finer market segmentation and target positioning; meanwhile, a real-time feedback circulation mechanism is introduced, and the behavior characteristic labels of the users are updated regularly according to the latest behavior mode and feedback data of the users. This dynamic update mechanism ensures the accuracy and timeliness of the user's tag.
S27, integrating the behavior feature labels of all users into a dynamically updated behavior feature database.
And integrating the behavior feature labels of all the users into a dynamically updated behavior feature database. This database provides a rich, real-time updated data resource for advertisement predictive models and personalized recommendations.
In this embodiment, it is specifically described that step S3 specifically includes:
S31, selecting a preset matching algorithm, and setting weights of the user behavior feature tags and the advertisement feature tags;
And selecting a proper preset matching algorithm and setting weights of the user behavior feature tag and the advertisement feature tag. The setting of the weights is critical because it determines the tag features that are emphasized in the matching process are more important, thereby affecting the scoring of the final degree of matching. The weight settings reflect the relative importance of the different features in the promotion policy.
S32, calculating the matching degree of each user behavior feature tag and each advertisement feature tag by using the selected matching algorithm; the matching degree calculating process is that a matching degree grading index is set to compare and obtain the matching degree (attraction and relativity) of each advertisement to the user;
calculating the matching degree of each user behavior feature tag and each advertisement feature tag by using the selected matching algorithm, and quantifying the attraction and the correlation degree of each advertisement to different users, namely the matching degree by setting a matching degree scoring index; this quantification process is critical to achieving personalized advertisement recommendations.
S33, analyzing a preliminary result of the matching degree evaluation operation, and identifying the user with the highest score to be paired with the advertisement;
Analyzing the preliminary result of the matching degree evaluation operation, and identifying the user with the highest score to be paired with the advertisement; these high scoring pairs represent optimal advertising opportunities, suggesting higher user engagement and advertising effectiveness.
S34, adjusting weights of the user behavior feature tags and the advertisement feature tags to conduct sensitivity analysis on the matching algorithm so as to evaluate the influence degree of tag weight change on matching degree scoring, and optimizing the matching algorithm;
Sensitivity analysis is carried out by adjusting weights of the user behavior feature tags and the advertisement feature tags, and the influence of tag weight change on matching degree scoring is estimated; through this step, it can be identified which weight adjustments can significantly improve the matching degree, and accordingly the matching algorithm is optimized, which helps to improve the sensitivity and accuracy of the matching algorithm.
S35, carrying out matching degree evaluation operation again after weight adjustment and optimization of a matching algorithm, and confirming a final matching degree score;
after weight adjustment and optimization of a matching algorithm, carrying out matching degree evaluation operation again, and confirming a final matching degree score; the method verifies the effect of the previous optimization measures, determines the matching degree score after optimization, and ensures the effectiveness of a matching algorithm.
S36, recording the final matching degree evaluation result, and transmitting the final matching degree evaluation result to the advertisement effect prediction model.
The final matching degree evaluation result is recorded and transmitted to the advertisement effect prediction model. This step ensures that the advertisement effectiveness prediction model operates based on the most accurate and optimal matching degree information, thereby improving the accuracy of the advertisement prediction model and the efficiency of advertisement delivery.
In this embodiment, it is specifically described that step S4 specifically includes:
S41, according to the matching degree evaluation result, calling an advertisement effect prediction model, and constructing a preliminary advertisement preview strategy frame through the advertisement effect prediction model;
the advertisement preview policy framework distributes the advertisement content to a user group with high matching degree according to the expected effect of the advertisement content, and preliminarily plans the display frequency, time period and format of the advertisement.
S42, adding user experience indexes into the primary advertisement preview policy framework to generate a primary advertisement preview policy;
The strategy for ensuring advertisement preview is not only based on matching degree and prediction effect, but also considers preference and acceptance degree of users.
The method comprises the steps of analyzing possible risk points in a preliminary advertisement preview strategy, such as potential objection of a user to certain advertisement contents, factors which change of market environment may influence advertisement effect and the like, and providing basis for strategy adjustment.
S43, setting an adjustment mechanism in the preliminary advertisement preview strategy, predefining an adjustment parameter, and starting the adjustment mechanism to adjust the preliminary advertisement preview strategy when the user feedback data triggers the adjustment parameter.
A flexible adjustment mechanism is set in the preliminary policy to quickly adjust the advertisement preview policy based on real-time data and feedback. This includes predefined tuning parameters (e.g., countermeasures when click-through rate drops to a certain level) to ensure effectiveness and timeliness of the policy.
And initially distributing advertisement delivery resources on the basis of considering the prediction effect and the user experience. This includes allocating a budget for advertising, selecting appropriate advertising channels and platforms, and planning production and distribution cycles for advertising content.
In this embodiment, it is specifically described that step S6 specifically includes:
s61, tracking and collecting feedback data of advertisement preview interaction of a user through a data collecting unit;
A data collection unit is built, which is the basis for the continuous optimization process, which tracks and collects feedback data of user interactions with the advertisement previews. The feedback data includes key indicators of click rate, viewing time, user messages, and score. The real-time monitoring system can be built to capture the behavior change of the user rapidly and provide real-time data support for further analysis.
S62, summarizing and sorting feedback data collected in real time, and taking the feedback data as an input variable of an optimized advertisement effect prediction model;
The feedback data collected in real time is summarized and arranged, so that the complicated data becomes orderly, and the method is easy to understand and analyze; the feedback data after the arrangement is used as an important input variable for optimizing the advertisement effect prediction model; this ensures that the advertisement predictive model can be adjusted according to the latest market and user response, enhancing the adaptability and predictability of the model.
S63, based on feedback data collected in real time, evaluating the current performance index of the advertisement effect prediction model, and formulating an optimization scheme according to the evaluation result of the performance index;
Based on the collected real-time feedback data, evaluating the current performance index of the advertisement effect prediction model; this includes evaluating key performance indicators of model accuracy, error rate. And according to the evaluation results of the performance indexes, formulating a targeted optimization scheme. The optimization scheme relates to one or more measures of adjusting the model structure, optimizing algorithm parameters and introducing new characteristic variables, and aims to solve the specific problems of the model.
S64, inputting feedback data collected in real time into the advertisement effect prediction model, and retraining the advertisement effect prediction model according to the optimization scheme and the feedback data so as to continuously optimize and update the advertisement effect prediction model.
Inputting feedback data collected in real time into an advertisement effect prediction model, and retraining the model according to a previously formulated optimization scheme; this step is critical to the continuous optimization process, which ensures that the model can learn and adapt based on the most current data, thus continuously improving the accuracy and effectiveness of the predictions. The retrained model is subjected to rigorous testing and validation and is used to replace or update the prior model only after ensuring stable performance and significant improvement.
Embodiment two:
The invention also provides an advertisement preview analysis system, which is used for realizing the advertisement preview analysis method as in the first embodiment; the advertisement preview analysis system specifically comprises:
the advertisement feature management unit is used for subdividing advertisement contents to generate advertisement feature labels;
The data collection unit is used for collecting historical browsing records, application using habits and feedback data of the user;
The user behavior analysis module is used for generating a user behavior characteristic tag according to the historical browsing record, the application using habit and the feedback data of the user;
The data processing unit is used for storing a preset matching algorithm and an advertisement effect prediction model; the preset matching algorithm is used for evaluating the matching degree of each user behavior feature tag and each advertisement feature tag, and the advertisement effect prediction model is used for generating a preliminary advertisement preview strategy according to the matching degree evaluation result;
the strategy optimization unit is used for continuously optimizing and updating the advertisement effect prediction model according to feedback data collected in real time;
And the monitoring unit comprises an interactive interface and is used for monitoring and presenting the running state of the advertisement preview analysis system.
Embodiment III:
The invention also provides a processor, which comprises a memory and at least one processor, wherein the memory stores instructions;
the processor invokes instructions in the memory to cause the processor to perform the method of advertisement preview analysis as in embodiment one.
Embodiment four:
The present invention also provides a storage medium, wherein instructions for implementing the method for advertisement preview analysis according to the first embodiment are stored on the storage medium.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A method of advertisement preview analysis, comprising:
according to the characteristics of the pre-placed advertisement content, subdividing the advertisement content to generate advertisement feature labels;
activating a user behavior analysis module by analyzing historical browsing records, application using habits and feedback data of a user, and generating a user behavior feature tag through the user behavior analysis module;
Evaluating the matching degree of each user behavior feature tag and each advertisement feature tag through a preset matching algorithm;
according to the matching degree evaluation result, invoking an advertisement effect prediction model to provide a preliminary advertisement preview strategy;
when a user interacts with advertisement previews, collecting feedback information of the user in real time through a data collecting unit;
continuously optimizing and updating the advertisement effect prediction model by utilizing feedback data collected in real time;
And optimizing the preliminary advertisement preview strategy based on the optimized and updated advertisement effect prediction model to generate corresponding advertisement putting strategy suggestions.
2. The method for analyzing advertisement previews according to claim 1, wherein the advertisement content is subdivided according to the characteristics of the pre-delivered advertisement content to generate advertisement feature labels; the method specifically comprises the following steps:
classifying the pre-placed advertisement content according to the main types of the advertisement content, and extracting and analyzing key attributes of each type of advertisement content;
Based on the key attributes of the advertisements, analyzing target groups, and analyzing to obtain user portraits expected to be touched by the advertisements;
associating the collected key attributes of the advertisement and the target group information with a specific consumption context;
and generating a preliminary advertisement feature label according to the classification of the advertisement content, the key attribute, the user portrait and the context associated information.
3. The method for analyzing advertisement previews according to claim 2, wherein the advertisement content is subdivided according to the characteristics of the pre-delivered advertisement content to generate advertisement feature tags; further comprises:
Verifying the generated preliminary advertisement feature labels in a small-scale target user group, and collecting preliminary feedback information;
and verifying and optimizing the preliminary advertisement feature labels through the preliminary feedback information to generate advertisement feature labels, and integrating the advertisement feature labels into a continuously updated label library.
4. A method of advertisement preview analysis according to claim 3, wherein the user behavior analysis module is activated by analyzing the user's historical browsing records, application usage habits and feedback data, and a user behavior feature tag is generated by the user behavior analysis module; the method specifically comprises the following steps:
collecting historical browsing records, application using habits and feedback data of historical advertisement contents of a user in a concentrated manner to obtain a first data packet;
performing preliminary cleaning processing on the first data packet, wherein the cleaning processing comprises the step of removing invalid or wrong data records;
identifying a main behavior pattern of a user from the cleaned first data packet by using a deep machine learning data mining technology;
deeply analyzing user feedback data of the user on the advertisement content to analyze preference and response intensity of the user on different types of advertisement content;
Activating a user behavior analysis module, and generating a group of user behavior feature labels for each user based on the identified main behavior mode and user feedback data;
Grouping users according to the generated behavior feature labels, and periodically updating the behavior feature labels according to the latest behavior mode of the users and user feedback data by using a real-time feedback circulation mechanism;
And integrating the behavior feature labels of all the users into a dynamically updated behavior feature database.
5. The method for analyzing advertisement previews according to claim 4, wherein the matching degree between each user behavior feature tag and each advertisement feature tag is evaluated through a preset matching algorithm; the method specifically comprises the following steps:
Selecting a preset matching algorithm, and setting weights of the user behavior feature tags and the advertisement feature tags;
Calculating the matching degree of each user behavior characteristic label and each advertisement characteristic label by using a selected matching algorithm;
analyzing the preliminary result of the matching degree evaluation operation, and identifying the user with the highest score to be paired with the advertisement;
The weights of the user behavior feature tags and the advertisement feature tags are adjusted to conduct sensitivity analysis on a matching algorithm, so that the influence degree of tag weight change on matching degree scoring is evaluated, and the matching algorithm is optimized;
After weight adjustment and optimization of a matching algorithm, carrying out matching degree evaluation operation again, and confirming a final matching degree score;
And recording the final matching degree evaluation result, and transmitting the final matching degree evaluation result to the advertisement effect prediction model.
6. The method of claim 5, wherein the advertisement preview method calls an advertisement effect prediction model to provide a preliminary advertisement preview strategy according to the matching degree evaluation result; the method specifically comprises the following steps:
According to the matching degree evaluation result, an advertisement effect prediction model is called, and a preliminary advertisement preview strategy framework is constructed through the advertisement effect prediction model;
adding user experience indexes into a preliminary advertisement preview policy framework to generate a preliminary advertisement preview policy;
Setting an adjustment mechanism in the preliminary advertisement preview strategy, predefining an adjustment parameter, and starting the adjustment mechanism to adjust the preliminary advertisement preview strategy when user feedback data triggers the adjustment parameter.
7. The method of advertisement preview analysis according to claim 1, wherein the advertisement effectiveness prediction model is continuously optimized and updated using feedback data collected in real time; the method specifically comprises the following steps:
tracking and collecting feedback data of advertisement preview interaction of a user through a data collecting unit;
Summarizing and sorting feedback data collected in real time, and taking the feedback data as an input variable of an optimized advertisement effect prediction model;
based on feedback data collected in real time, evaluating the current performance index of the advertisement effect prediction model, and formulating an optimization scheme according to the evaluation result of the performance index;
and inputting feedback data collected in real time into the advertisement effect prediction model, and retraining the advertisement effect prediction model according to an optimization scheme and the feedback data so as to continuously optimize and update the advertisement effect prediction model.
8. An advertisement preview analysis system, characterized by a method for implementing the advertisement preview analysis of any one of claims 1 to 7; the advertisement preview analysis system specifically comprises:
the advertisement feature management unit is used for subdividing advertisement contents to generate advertisement feature labels;
The data collection unit is used for collecting historical browsing records, application using habits and feedback data of the user;
The user behavior analysis module is used for generating a user behavior characteristic tag according to the historical browsing record, the application using habit and the feedback data of the user;
The data processing unit is used for storing a preset matching algorithm and an advertisement effect prediction model; the preset matching algorithm is used for evaluating the matching degree of each user behavior characteristic label and each advertisement characteristic label, and the advertisement effect prediction model is used for generating a preliminary advertisement preview strategy according to the matching degree evaluation result;
the strategy optimization unit is used for continuously optimizing and updating the advertisement effect prediction model according to feedback data collected in real time;
and the monitoring unit comprises an interactive interface and is used for monitoring and presenting the running state of the advertisement preview analysis system.
9. A processor comprising a memory and at least one processor, the memory having instructions stored therein;
the processor invoking the instructions in the memory to cause the processor to perform the method of advertisement preview analysis of any of claims 1 to 7.
10. A storage medium having stored thereon instructions for implementing the method of advertisement preview analysis of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410300980.0A CN118071427A (en) | 2024-03-15 | 2024-03-15 | Advertisement preview analysis method, system, processor and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410300980.0A CN118071427A (en) | 2024-03-15 | 2024-03-15 | Advertisement preview analysis method, system, processor and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118071427A true CN118071427A (en) | 2024-05-24 |
Family
ID=91097201
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410300980.0A Pending CN118071427A (en) | 2024-03-15 | 2024-03-15 | Advertisement preview analysis method, system, processor and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118071427A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106803190A (en) * | 2017-01-03 | 2017-06-06 | 北京掌阔移动传媒科技有限公司 | A kind of ad personalization supplying system and method |
| CN111882362A (en) * | 2020-07-31 | 2020-11-03 | 苏州云开网络科技有限公司 | Artificial intelligence advertisement delivery system based on 5G communication network |
| CN116843388A (en) * | 2023-08-29 | 2023-10-03 | 新义互联(北京)科技有限公司 | Advertisement delivery analysis method and system |
| CN116993417A (en) * | 2023-08-16 | 2023-11-03 | 安徽盟弘信息科技有限公司 | Intelligent advertisement popularization platform |
| CN117252647A (en) * | 2023-09-21 | 2023-12-19 | 深圳市聚海通达科技有限公司 | Digitalized integrated marketing service system |
| CN117593053A (en) * | 2023-11-02 | 2024-02-23 | 深圳市生苦科技有限公司 | Big data analysis system for advertisement delivery |
-
2024
- 2024-03-15 CN CN202410300980.0A patent/CN118071427A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106803190A (en) * | 2017-01-03 | 2017-06-06 | 北京掌阔移动传媒科技有限公司 | A kind of ad personalization supplying system and method |
| CN111882362A (en) * | 2020-07-31 | 2020-11-03 | 苏州云开网络科技有限公司 | Artificial intelligence advertisement delivery system based on 5G communication network |
| CN116993417A (en) * | 2023-08-16 | 2023-11-03 | 安徽盟弘信息科技有限公司 | Intelligent advertisement popularization platform |
| CN116843388A (en) * | 2023-08-29 | 2023-10-03 | 新义互联(北京)科技有限公司 | Advertisement delivery analysis method and system |
| CN117252647A (en) * | 2023-09-21 | 2023-12-19 | 深圳市聚海通达科技有限公司 | Digitalized integrated marketing service system |
| CN117593053A (en) * | 2023-11-02 | 2024-02-23 | 深圳市生苦科技有限公司 | Big data analysis system for advertisement delivery |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10846617B2 (en) | Context-aware recommendation system for analysts | |
| Besbes et al. | Optimization in online content recommendation services: Beyond click-through rates | |
| US10096040B2 (en) | Management of the display of online ad content consistent with one or more performance objectives for a webpage and/or website | |
| CN102708131B (en) | By consumer's automatic classification in fine point | |
| US11436434B2 (en) | Machine learning techniques to identify predictive features and predictive values for each feature | |
| US20160210658A1 (en) | Determining touchpoint attributions in a segmented media campaign | |
| US20140236708A1 (en) | Methods and apparatus for a predictive advertising engine | |
| US20170323330A1 (en) | Media spend management using real-time predictive modeling of touchpoint exposure effects | |
| EP3326136A1 (en) | Sequential delivery of advertising content across media devices | |
| CN106250532A (en) | Application recommendation method, device and server | |
| CN106447419B (en) | Visitor identification based on feature selection | |
| Klašnja-Milićević et al. | Recommender systems in e-learning environments | |
| US20200302486A1 (en) | Method and system for determining optimized customer touchpoints | |
| US11720808B2 (en) | Feature removal framework to streamline machine learning | |
| US20190034843A1 (en) | Machine learning system and method of grant allocations | |
| Wang et al. | Probabilistic models for ad viewability prediction on the web | |
| CN118313884B (en) | Personalized information sending method and device | |
| CN118154031B (en) | Multi-mode entrepreneur project product uniqueness assessment and product improvement method | |
| CN117291655A (en) | Consumer life cycle operation analysis method based on entity and network collaborative mapping | |
| CN118071427A (en) | Advertisement preview analysis method, system, processor and storage medium | |
| CN117808480A (en) | Multi-channel interactive customer relationship management system | |
| CN117271905A (en) | Crowd image-based lateral demand analysis method and system | |
| WO2020106950A1 (en) | User-experience development system | |
| CN112200602B (en) | Neural network model training method and device for advertisement recommendation | |
| CN116485352B (en) | Member management and data analysis method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |