Advertisement delivery optimization method and system based on big data
Through big data technology matching advertising attribute information and delivery algorithms, dynamically adjusting advertising delivery strategies, solving the problem of inaccurate advertising delivery in the existing technology and improving advertising promotion effect.
Patent Information
- Application Number
- CN202410942041.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-07-15
AI Technical Summary
The existing advertising delivery technology cannot achieve precise advertising delivery, resulting in low advertising effectiveness.
By obtaining the target attribute information of the target advertisement, using big data technology to match the advertising attribute information in the advertising database with the target attribute information, selecting the appropriate advertising delivery algorithm and control parameters, and dynamically adjusting the delivery strategy to improve advertising effectiveness.
It has achieved the accuracy and effect of advertising delivery, stronger adaptability, conformed to the current actual situation, and improved the advertising effect.
Smart Images

Figure CN118469646B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of big data technology, and in particular to an advertising delivery optimization method and system based on big data. Background Art
[0002] With the rapid development of electronic technology and network technology, electronic devices (such as mobile phones, tablet computers, etc.) are becoming more and more popular, and advertising is also developing rapidly in electronic devices. At present, advertising is simple and crude, and it is impossible to accurately achieve advertising, which reduces the effectiveness of advertising. Therefore, how to achieve accurate advertising to improve the effectiveness of advertising needs to be solved urgently. Summary of the invention
[0003] The embodiments of the present application provide an advertising delivery optimization method and system based on big data, which can improve the advertising effect.
[0004] In a first aspect, an embodiment of the present application provides an advertising delivery optimization method based on big data, the method comprising:
[0005] Obtain target attribute information of the target advertisement;
[0006] Using big data technology, the attribute information of Q advertisements in the advertisement database is matched with the target attribute information to obtain P attribute information that successfully matches the target attribute information, and reference advertisements corresponding to the P attribute information are obtained to obtain P reference advertisements, where each advertisement in the Q advertisements corresponds to one attribute information, where P is an integer greater than 1, and Q is an integer greater than or equal to P;
[0007] Obtaining advertisement delivery algorithms corresponding to the P reference advertisements to obtain M advertisement delivery algorithms, where M is a positive integer less than or equal to P;
[0008] Selecting a target advertisement delivery algorithm from the M advertisement delivery algorithms;
[0009] Obtaining a first algorithm control parameter corresponding to the target advertisement delivery algorithm;
[0010] Delivering the target advertisement according to the first algorithm control parameter and the target advertisement delivery algorithm;
[0011] Obtaining target feedback information for the target advertisement;
[0012] Optimize the first algorithm control parameter according to the target feedback information to obtain the second algorithm control parameter;
[0013] The target advertisement is delivered according to the second algorithm control parameter and the target advertisement delivery algorithm to improve the delivery effect of the target advertisement.
[0014] In a second aspect, the embodiment of the present application provides an advertising delivery optimization system based on big data, wherein the system comprises: an acquisition unit, a selection unit, a delivery unit, and an optimization unit, wherein:
[0015] The acquisition unit is used to acquire target attribute information of a target advertisement; use big data technology to match the attribute information of Q advertisements in an advertisement database with the target attribute information to obtain P attribute information that successfully matches the target attribute information, acquire reference advertisements corresponding to the P attribute information to obtain P reference advertisements, each of the Q advertisements corresponds to one attribute information, P is an integer greater than 1, and Q is an integer greater than or equal to P; acquire advertisement delivery algorithms corresponding to the P reference advertisements to obtain M advertisement delivery algorithms, where M is a positive integer less than or equal to P;
[0016] A selection unit, configured to select a target advertisement delivery algorithm from the M advertisement delivery algorithms;
[0017] The acquisition unit is further used to acquire a first algorithm control parameter corresponding to the target advertisement delivery algorithm;
[0018] The delivery unit is configured to deliver the target advertisement according to the first algorithm control parameter and the target advertisement delivery algorithm;
[0019] The acquisition unit is further used to acquire target feedback information for the target advertisement;
[0020] The optimization unit is used to optimize the first algorithm control parameter according to the target feedback information to obtain the second algorithm control parameter;
[0021] The delivery unit is further used to deliver the target advertisement according to the second algorithm control parameter and the target advertisement delivery algorithm to improve the delivery effect of the target advertisement.
[0022] In a third aspect, an embodiment of the present application provides a server, comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps in the first aspect of the embodiment of the present application.
[0023] The implementation of the embodiments of the present application has the following beneficial effects:
[0024] It can be seen that the big data-based advertising delivery optimization method and system described in the embodiments of the present application obtain target attribute information of a target advertisement, use big data technology to match the attribute information of Q advertisements in an advertising database with the target attribute information, obtain P attribute information that successfully matches the target attribute information, obtain reference advertisements corresponding to the P attribute information, obtain P reference advertisements, each of the Q advertisements corresponds to one attribute information, P is an integer greater than 1, Q is an integer greater than or equal to P, obtain advertising delivery algorithms corresponding to the P reference advertisements, obtain M advertising delivery algorithms, M is a positive integer less than or equal to P, select a target advertising delivery algorithm from the M advertising delivery algorithms, obtain a first algorithm control parameter corresponding to the target advertising delivery algorithm, and select the first algorithm control parameter according to the first algorithm control parameter. And the target advertisement delivery algorithm delivers the target advertisement, obtains the target feedback information for the target advertisement, optimizes the first algorithm control parameter according to the target feedback information, obtains the second algorithm control parameter, delivers the target advertisement according to the second algorithm control parameter and the target advertisement delivery algorithm, so as to improve the delivery effect of the target advertisement. In this way, big data can be used to select a target advertisement delivery algorithm that is adapted to its own attributes for the target advertisement, and delivers the target advertisement with the first algorithm control parameter under the target advertisement delivery algorithm. The algorithm control parameters of the target advertisement delivery algorithm can also be adjusted based on the dynamic feedback of the delivery effect of the target advertisement, so that the optimized advertising effect is further improved and more in line with the current actual situation, so that the target advertisement delivery algorithm can deliver the target advertisement more adaptively, thereby improving the advertising publicity effect. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0026] Figure 1 It is a flowchart of an advertising delivery optimization method based on big data provided in an embodiment of the present application;
[0027] Figure 2 It is a structural diagram of a server provided in an embodiment of the present application;
[0028] Figure 3 It is a block diagram of the functional units of a big data-based advertising delivery optimization system provided in an embodiment of the present application. DETAILED DESCRIPTION
[0029] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0030] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.
[0031] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0032] In an embodiment of the present application, the advertising delivery algorithm may include at least one of the following: a content-based recommendation algorithm, a user-based recommendation algorithm, a collaborative filtering-based recommendation algorithm, a deep learning-based recommendation algorithm, a graph convolutional network-based recommendation algorithm, a Bayesian network-based recommendation algorithm, a multi-layer perceptron network-based recommendation algorithm, a popularity-based recommendation algorithm, etc., without limitation here, and different advertising delivery algorithms may correspond to different algorithm control parameters.
[0033] The embodiments of the present application are described in detail below.
[0034] See also Figure 1 , Figure 1 1 is a flow chart of an advertising delivery optimization method based on big data provided in an embodiment of the present application. As shown in the figure, the advertising delivery optimization method based on big data includes:
[0035] 101. Obtain target attribute information of the target advertisement.
[0036] In the embodiment of the present application, the target attribute information may include at least one of the following: advertisement type, advertisement purpose, advertisement subject matter, advertisement duration, advertisement theme, advertisement producer, advertisement budget, advertisement time period, advertisement image effect, advertisement spokesperson, etc., which are not limited here. In a specific implementation, the target attribute information of the target advertisement may be obtained.
[0037] 102. Use big data technology to match the attribute information of Q advertisements in the advertisement database with the target attribute information to obtain P attribute information that successfully matches the target attribute information, obtain reference advertisements corresponding to the P attribute information, and obtain P reference advertisements, where each of the Q advertisements corresponds to one attribute information, P is an integer greater than 1, and Q is an integer greater than or equal to P.
[0038] The advertisement database may pre-store multiple advertisements, for example, Q advertisements, where Q is a positive integer.
[0039] In a specific implementation, big data technology can be used to match the attribute information of Q advertisements in the advertisement database with the target attribute information, and obtain P attribute information that successfully matches the target attribute information, obtain reference advertisements corresponding to the P attribute information, and obtain P reference advertisements. Each of the Q advertisements corresponds to one attribute information, P is an integer greater than 1, and Q is an integer greater than or equal to P. In this way, big data technology can be used to find advertisements similar to the target advertisement, so as to adapt the best advertising delivery algorithm based on these similar advertisements to ensure the effectiveness of the advertising delivery.
[0040] 103. Obtain advertisement delivery algorithms corresponding to the P reference advertisements to obtain M advertisement delivery algorithms, where M is a positive integer less than or equal to P.
[0041] In the embodiment of the present application, since each of the P reference advertisements corresponds to an advertising delivery algorithm, but different advertisements may use the same advertising delivery algorithm, therefore, the advertising delivery algorithms corresponding to the P reference advertisements can be obtained to obtain M advertising delivery algorithms, where M is a positive integer less than or equal to P.
[0042] 104. Select a target advertisement delivery algorithm from the M advertisement delivery algorithms.
[0043] In a specific implementation, at least one advertisement delivery algorithm may be selected from the M advertisement delivery algorithms as a target advertisement delivery algorithm.
[0044] In a specific implementation, the advertising delivery algorithms corresponding to P reference advertisements may be obtained to obtain M advertising delivery algorithms, where M is a positive integer less than or equal to P. A target advertising delivery algorithm may be selected from the M advertising delivery algorithms. The target advertising delivery algorithm may be one of the M advertising delivery algorithms. For example, the target advertising delivery algorithm may be the advertising delivery algorithm with the highest algorithm score among the M advertising delivery algorithms.
[0045] Optionally, the above step 104, selecting a target advertisement delivery algorithm from the M advertisement delivery algorithms, may include the following steps:
[0046] 41. Obtain push result information and algorithm evaluation parameters of each of the M advertisement delivery algorithms to obtain m push result information and m algorithm evaluation parameters;
[0047] 42. Determine a first evaluation value corresponding to each of the m push result information to obtain m first evaluation values;
[0048] 43. Determine a second evaluation value corresponding to each of the m algorithm evaluation parameters to obtain m second evaluation values;
[0049] 44. Perform a weighted operation according to the m first evaluation values and the m second evaluation values to obtain m target evaluation values;
[0050] 45. Select a maximum value among the m target evaluation values, and determine the advertisement delivery algorithm corresponding to the maximum value as the target advertisement delivery algorithm.
[0051] In the specific implementation, the push result information of each of the M advertising delivery algorithms and the algorithm evaluation parameters can be obtained, and m push result information and m algorithm evaluation parameters can be obtained. Each advertising delivery algorithm corresponds to a push result information and an algorithm evaluation parameter. Among them, the push result information is used to reflect the advertising push effect, which mainly reflects the user's subjective evaluation of the advertising content, and indirectly reflects the objective evaluation of the advertising delivery algorithm. The algorithm evaluation parameters are used to evaluate the quality of the advertising delivery algorithm itself, that is, they reflect the subjective evaluation of the advertising delivery algorithm.
[0052] Furthermore, the mapping relationship between the preset push result information and the first evaluation value can be pre-stored, and then, the first evaluation value corresponding to each push result information in the m push result information can be determined based on the mapping relationship to obtain m first evaluation values. The mapping relationship between the preset algorithm evaluation parameters and the second evaluation value can also be pre-stored, and then, the second evaluation value corresponding to each algorithm evaluation parameter in the m algorithm evaluation parameters can be determined based on the mapping relationship to obtain m second evaluation values.
[0053] In a specific implementation, different advertising delivery algorithms may correspond to different weight pairs, and the weight pairs may be empirical values. Specifically, a mapping relationship between a preset advertising delivery algorithm and a weight pair may be pre-set to determine the weight pair corresponding to each advertising delivery algorithm based on the mapping relationship. The weight pair includes a first weight and a second weight. The first weight is the weight corresponding to the first evaluation value, and the second weight is the weight corresponding to the second evaluation value. The first weight + the second weight = 1. Furthermore, a weighted operation may be performed based on the m first evaluation values and the m second evaluation values and their corresponding weight pairs to obtain m target evaluation values, and then the m target evaluation values are selected. The maximum value in , and the advertising delivery algorithm corresponding to the maximum value is determined as the target advertising delivery algorithm. In this way, the push result information is used to reflect the advertising push effect, which mainly reflects the user's subjective evaluation of the advertising content, and also reflects the objective evaluation of the advertising delivery algorithm. The algorithm evaluation parameters are used to evaluate the quality of the advertising delivery algorithm itself, that is, it reflects the subjective evaluation of the advertising delivery algorithm. The push result information and algorithm evaluation parameters of the advertising delivery algorithm can be used to evaluate the quality of an advertising delivery algorithm from a subjective and objective perspective. Then, the advertising delivery algorithm that best matches the target advertisement can be selected, which helps to ensure the efficiency of advertising delivery in the later stage.
[0054] Optionally, the above step 42, determining the first evaluation value corresponding to each of the m push result information to obtain the m first evaluation values, may include the following steps:
[0055] 421. Obtain first push result information, the first push result information including at least one piece of push result information of an advertisement, each piece of push result information including: the number of potential push objects, the number of actual push objects, and advertisement browsing status; the first push result information is any one of the m pieces of push result information; the advertisement browsing status includes at least one of the following: browsing time, browsing attention content, and content time period for browsing advertisement content;
[0056] 422. Determine a first ratio between the actual number of push objects and the potential number of push objects in the push result information of each advertisement in the push result information of the at least one advertisement, and determine an advertisement browsing index factor according to the advertisement browsing situation;
[0057] 423. Determine at least one evaluation value according to the first ratio of each advertisement in the at least one advertisement and the advertisement browsing index factor;
[0058] 424. Acquire attribute information of the at least one advertisement to obtain at least one attribute information;
[0059] 425. Determine a weight corresponding to each attribute information in the at least one attribute information to obtain at least one weight;
[0060] 426. Perform a weighted operation on the at least one weight and the at least one evaluation value to obtain the first evaluation value.
[0061] In a specific implementation, first push result information may be obtained, the first push result information including push result information of at least one advertisement, the push result information of the at least one advertisement may be understood as push result information of an advertisement previously delivered by the advertisement delivery algorithm, each push result information including: the number of potential push objects, the number of actual push objects, and advertisement browsing status, the first push result information is any push result information of the m push result information, wherein the number of potential push objects may be understood as the number of push objects locked by the target advertisement delivery algorithm, the number of actual push objects may be understood as the push objects to which the target advertisement is actually pushed by the target advertisement delivery algorithm and which meet preset conditions, the preset conditions may be pre-set or system defaulted, the preset conditions are to ensure that the target advertisement reaches the user normally and the user confirms that the advertisement has been viewed, in actual applications, the target advertisement may be actually pushed due to the user swiping the screen, refreshing too fast, the device running out of power, system upgrade, software upgrade, etc., but the user terminal does not receive or has no time to view the target advertisement in the future, that is, these push objects may be removed through the preset conditions, and the actual number of push objects may be obtained.
[0062] Among them, the advertisement browsing situation may include at least one of the following: browsing time, browsing attention content, browsing advertising content content time period, etc., which are not limited here. The advertisement browsing index factor reflects the user's perception of the target advertisement to a certain extent. Among them, browsing time can be understood as the overall time of watching the advertisement, browsing attention content can be understood as the focus of attention in the advertisement. Generally, browsing attention content may reduce the playback rate or play it repeatedly, and browsing advertising content content time period can be understood as the core value coverage of the advertisement.
[0063] In a specific implementation, for each user, the user's ad browsing situation can be obtained, and the ad browsing situation may include: browsing time, browsing focus content, and content time period for browsing ad content. Then the ratio between browsing time and ad time can be determined to obtain the browsing completeness of the ad. The correlation between the browsing focus content and the core content of the ad can be determined. The higher the correlation, the greater the ad browsing index factor. Determine the content time period where the core content of the ad is located, determine the intersection content time period between the content time period for browsing the ad content and the content time period where the core content of the ad is located, determine the ratio between the intersection content time period and the content time period where the core content of the ad is located, and obtain the core value influence of the ad. The greater the core value influence of the ad, the greater the ad browsing index factor.
[0064] Furthermore, different advertisements have different proportions of core contents. Thus, the first weight and the second weight can be determined based on the proportion of the core contents. For example, a mapping relationship between a preset correlation and an advertisement browsing index factor can be pre-stored, and a first advertisement browsing index factor corresponding to the correlation between the browsing attention content and the core content of the advertisement can be determined based on the mapping relationship. A mapping relationship between a preset core value influence and an advertisement browsing index factor can also be pre-stored, and a second advertisement browsing index factor corresponding to the core value influence of the advertisement can be determined based on the mapping relationship. The first weight corresponds to the first advertisement browsing index factor, and the second weight corresponds to the second advertisement browsing index factor. Then, based on the first weight, the second weight, the first advertisement browsing index factor and the second advertisement browsing index factor, a first weight is calculated. The ad browsing index factor is weighted and calculated to obtain the reference ad browsing index factor. The browsing completeness reflects the overall quality of the ad and the effect of the ad delivery. Furthermore, different browsing completenesses may correspond to different impact coefficients, i.e., the greater the browsing completeness, the greater the impact coefficient. The mapping relationship between the preset browsing completeness and the impact coefficient may be pre-stored. Furthermore, the target impact coefficient corresponding to the ad browsing completeness may be determined based on the mapping relationship. The target ad browsing index factor = reference ad browsing index factor * (1 + target impact coefficient). In this way, the ad delivery effect may be fed back based on the degree of user browsing of the ad and the depth of the content they pay attention to. That is, only when the ad is delivered to the right person can the maximum advertising effect be achieved, which also indirectly feeds back the accuracy of the ad delivery.
[0065] In a specific implementation, a first ratio between the actual number of push objects and the potential number of push objects in the push result information of each advertisement in the push result information of at least one advertisement can be determined. The first ratio reflects the success rate of advertisement push to a certain extent, and an advertisement browsing index factor is determined according to the advertisement browsing situation. The advertisement browsing index factor reflects the user perception of the target advertisement. At least one evaluation value is then determined based on the first ratio of each advertisement in at least one advertisement and the advertisement browsing index factor. The evaluation value of the advertisement is not only for the advertisement, but also for the advertisement delivery algorithm to a certain extent. This can help to make subsequent in-depth decisions on a target advertisement delivery algorithm that is suitable for the target advertisement, based on in-depth consideration of the delivery accuracy, advertisement delivery success rate and user feedback of the advertisement delivery algorithm.
[0066] Next, the attribute information of at least one advertisement may be obtained to obtain at least one attribute information, and then a weight corresponding to each attribute information in the at least one attribute information may be determined to obtain at least one weight. Specifically, the at least one attribute information may be matched with the target attribute information to obtain at least one matching value. Different matching values may correspond to different weights, that is, the sum of at least one matching value may be determined, and then a ratio between each matching value in the at least one matching value and the sum may be determined to obtain at least one weight. Then, a weighted operation may be performed based on the at least one weight and the at least one evaluation value to obtain a first evaluation value.
[0067] Optionally, the above step 423, determining at least one evaluation value according to the first ratio of each advertisement in the at least one advertisement and the advertisement browsing index factor, may include the following steps:
[0068] A1. Obtain a target first ratio and a target advertisement browsing index factor for a first advertisement, where the first advertisement is any advertisement among the at least one advertisement;
[0069] A2. determining a target adjustment parameter corresponding to the target first ratio;
[0070] A3. Determine a reference evaluation value corresponding to the target advertisement browsing index factor;
[0071] A4. Obtain a target first score for the first advertisement, where the target first score is related to a preset operation for the first advertisement, where the preset operation includes at least one of the following: like, favorite, comment, and forward;
[0072] A5. Determine a target fine-tuning parameter corresponding to the target first score;
[0073] A6. Adjust the reference evaluation value according to the target adjustment parameter and the target fine-tuning parameter to obtain an evaluation value of the first advertisement.
[0074] In a specific implementation, taking the first advertisement as an example, the first advertisement is any advertisement among at least one advertisement, and the target first ratio of the first advertisement and the target advertisement browsing index factor can be obtained. The mapping relationship between the preset first ratio and the adjustment parameter can also be pre-stored, and then the target adjustment parameter corresponding to the target first ratio of the first advertisement can be determined.
[0075] In a specific implementation, a mapping relationship between preset advertisement browsing index factors and evaluation values may be pre-stored, and then a reference evaluation value corresponding to the target advertisement browsing index factor of the first advertisement may be determined based on the mapping relationship.
[0076] Next, the target first score of the first advertisement can also be obtained. The target first score can be achieved by user scoring. Of course, the target first score can also be related to the first advertisement and a preset operation, wherein the preset operation can include at least one of the following: like, favorite, comment, forward, etc., which are not limited here.
[0077] In a specific implementation, a mapping relationship between a preset first score and a fine-tuning parameter can be pre-stored, and then, based on the mapping relationship, a target fine-tuning parameter corresponding to the target first score can be determined, and then the reference evaluation value can be adjusted according to the target adjustment parameter and the target fine-tuning parameter to obtain an evaluation value of the first advertisement, as follows: Evaluation value of the first advertisement = (1 + target adjustment parameter) * (1 + target fine-tuning parameter) * reference evaluation value. The first ratio reflects the success rate of advertisement push to a certain extent, and the advertisement browsing index factor reflects the user perception of the advertisement. On the one hand, the advertisement can be preliminarily evaluated based on the user perception of the target advertisement. Usually, different advertisements are suitable for different classes and ages, that is, different advertisements correspond to different preference groups, and different preference groups Different users have different perceptions of advertisements. Therefore, user perception reflects the delivery accuracy of the ad delivery algorithm to a certain extent, and adjusts the evaluation results of its preliminary evaluation based on its ad push success rate, which can deeply optimize the delivery accuracy of the ad delivery algorithm. On the other hand, the ad rating reflects the response of the ad and the ad delivery accuracy, and can further optimize the delivery accuracy of the ad delivery algorithm and the response of the ad based on the ad rating. The evaluation value of the ad is not only for the ad, but also for the ad delivery algorithm to a certain extent, so that on the basis of in-depth consideration of the ad delivery accuracy, ad delivery success rate and user feedback of the ad delivery algorithm, a target ad delivery algorithm suitable for the target ad can be deeply decided.
[0078] 105. Obtain a first algorithm control parameter corresponding to the target advertisement delivery algorithm.
[0079] In the embodiments of the present application, different advertising delivery algorithms may correspond to different algorithm control parameters, and the algorithm control parameters may be used to control the delivery effect, delivery area, delivery frequency, delivery location, delivery time, delivery form, etc. of the advertising delivery algorithm, which are not limited here.
[0080] In a specific implementation, different advertising delivery algorithms may correspond to different algorithm control parameters. Thus, a first algorithm control parameter corresponding to a target advertising delivery algorithm may be obtained, and then the target advertisement may be delivered according to the first algorithm control parameter and the target advertising delivery algorithm.
[0081] In a specific implementation, a first algorithm control parameter corresponding to the target advertisement delivery algorithm may be obtained.
[0082] 106. Deliver the target advertisement according to the first algorithm control parameter and the target advertisement delivery algorithm.
[0083] In the embodiment of the present application, the target advertisement can be delivered according to the first algorithm control parameters and the target advertisement delivery algorithm. In this way, an adaptive advertisement delivery algorithm can be selected for advertisement delivery, which can preliminarily guarantee the advertisement delivery effect to a certain extent.
[0084] Optionally, the above step 106, placing the target advertisement according to the first algorithm control parameter and the target advertisement placing algorithm, may include the following steps:
[0085] 61. Determine k delivery objects according to the first algorithm control parameter and the target advertisement delivery algorithm, where k is a positive integer;
[0086] 62. Obtain habit parameters of a target user of a first delivery object; the first delivery object is any delivery object among the k delivery objects;
[0087] 63. Determine target push parameters of the target advertisement according to the target user habit parameters;
[0088] 64. Push the target advertisement to the first delivery object according to the target push parameters.
[0089] Among them, the user habit parameters of the first delivery object may include at least one of the following: the preferred advertising type of the first delivery object, the time period of the first delivery object's attention to advertising, the consumption situation of the first delivery object, etc., which are not limited here.
[0090] The target push parameters may include at least one of the following: push time, advertisement duration, advertisement play rate, advertisement preference parameters, etc., which are not limited here.
[0091] In a specific implementation, k delivery objects can be determined according to the first algorithm control parameter and the target advertisement delivery algorithm, where k is a positive integer, that is, different advertisement delivery algorithms can lock some push objects based on their algorithms and corresponding algorithm control parameters.
[0092] Next, the target user habit parameters of the first delivery object can also be obtained. The first delivery object is any one of the k delivery objects, that is, the corresponding push parameters are determined for each locked push object, that is, the target advertisement is dynamically pushed based on the user habit parameters, and the advertisement playback effect can also be dynamically adjusted based on the target push parameters, so that the advertisement can deeply conform to the user habits, that is, the target advertisement is played for the user at the best time with the best effect, and the advertisement push effect is deeply improved.
[0093] 107. Obtain target feedback information for the target advertisement.
[0094] The target feedback information may include at least one of the following: a score for the target advertisement, a comment on the target advertisement, a forwarding of the target advertisement, an ignoring of the target advertisement, a blacklisting of the target advertisement, etc., which are not limited here. Then, the target feedback information for the target advertisement may be obtained.
[0095] 108. Optimize the first algorithm control parameter according to the target feedback information to obtain a second algorithm control parameter.
[0096] Among them, the target feedback information to a certain extent indicates that there are some shortcomings in the target advertisement or the target advertisement delivery algorithm. The target feedback information can be used to adjust the first algorithm control parameters to make up for the shortcomings of the advertisement itself and optimize the delivery effect of the target advertisement delivery algorithm itself, which will help to further improve the delivery effect of the target advertisement delivery algorithm and enhance the user perception of the target advertisement.
[0097] In a specific implementation, a mapping relationship between preset feedback information and optimization parameters can be pre-stored, and then, based on the mapping relationship, the target optimization parameters corresponding to the target feedback information can be determined, and the first algorithm control parameters can be optimized using the target optimization parameters to obtain the second algorithm control parameters. The target advertisement is delivered according to the second algorithm control parameters and the target advertisement delivery algorithm to improve the delivery effect of the target advertisement. In this way, the corresponding algorithm control parameters can be dynamically adjusted according to the advertisement feedback situation, and then, the delivery effect of the target advertisement can be improved.
[0098] Optionally, the target feedback information includes target statistical information of designated operations for the target advertisement within a preset time period, and the designated operations include: like, favorite, comment, forward, download, blacklist, and dislike; the above step 108, optimizing the first algorithm control parameter according to the target feedback information to obtain the second algorithm control parameter, may include the following steps:
[0099] 81. Determine a target advertisement popularity according to the target statistical information and the length of the preset time period;
[0100] 82. Determine the positive evaluation result statistical information and the negative evaluation result statistical information of the target advertisement according to the target statistical information;
[0101] 83. Determine a target ratio between the positive evaluation result statistical information and the negative evaluation result statistical information;
[0102] 84. Obtain a target reference ratio corresponding to the target advertisement popularity;
[0103] 85. Determine a target deviation value between the target ratio and the target reference ratio;
[0104] 86. Determine a target feedback optimization coefficient corresponding to the target deviation;
[0105] 87. Optimize the first algorithm control parameter according to the target feedback optimization coefficient to obtain the second algorithm control parameter.
[0106] In a specific implementation, a positive evaluation result may include at least one of the following: likes, favorites, positive comment results semantically represented in comments, forwarding, downloading, etc., without limitation here; a negative evaluation result may include at least one of the following: negative evaluation results semantically represented in comments, blacklisting, dislike, etc., without limitation here.
[0107] The preset time period may be preset or set by system default. The preset time period may refer to a part or all of a period of time after the target advertisement is delivered by the target advertisement delivery algorithm.
[0108] In an embodiment of the present application, the target advertisement popularity can be determined based on the target statistical information and the length of a preset time period. The target statistical information reflects the user attention level of the target advertisement. Furthermore, the target advertisement popularity can be determined using the target statistical information and the length of a preset time period. The more times users browse, the greater the advertisement popularity. Alternatively, the more times users comment, the greater the advertisement popularity. Alternatively, the more times users forward, the greater the advertisement popularity. Alternatively, the more times users collect, the greater the advertisement popularity. Furthermore, the target advertisement popularity can be determined based on the statistical information of each dimension. For example, the target advertisement popularity within the length of a preset time period can be counted. Of course, the advertisement popularity within a time period can also be counted. The ratio between the advertisement popularity and the length of the time period can also be referred to as the advertisement popularity per unit time.
[0109] In a specific implementation, the positive evaluation result statistical information and the negative evaluation result statistical information of the target advertisement can be determined according to the target statistical information, and the target ratio between the positive evaluation result statistical information and the negative evaluation result statistical information can be determined. The ratio between the two reflects the degree of positive impact of the advertisement.
[0110] Furthermore, a mapping relationship between preset advertisement heat and reference ratio may be pre-stored, that is, different advertisement heats correspond to different degrees of positive impact, and a target reference ratio corresponding to a target advertisement heat may be obtained based on the mapping relationship.
[0111] Next, the target deviation value between the target ratio and the target reference ratio can be determined. The target deviation value = (target ratio - target reference ratio) / target reference ratio. The deviation reflects to a certain extent whether the social influence of the advertisement is positive or negative. If it is negative, in order to reduce the social influence, the advertising popularity can be restrained to a certain extent. Conversely, if it is positive, in order to enhance the social influence, the advertising popularity can be enhanced to a certain extent.
[0112] In a specific implementation, a mapping relationship between a preset deviation and a feedback optimization coefficient can be pre-stored, and a target feedback optimization coefficient corresponding to the target deviation can be determined based on the mapping relationship. Then, the first algorithm control parameter can be optimized according to the target feedback optimization coefficient to obtain a second algorithm control parameter, i.e., the second algorithm control parameter = (1 + first algorithm control parameter) * target feedback optimization coefficient. That is, through various statistics of advertisements, the social influence of advertisements can be deeply fed back, positive energy can be promoted, negative energy can be suppressed, and the effect of advertisement delivery can be dynamically adjusted, so that the depth of advertisement delivery is in line with the public's wishes and the effect of advertisement delivery can be guaranteed to a certain extent.
[0113] 109. Deliver the target advertisement according to the second algorithm control parameter and the target advertisement delivery algorithm to improve delivery effect of the target advertisement.
[0114] In a specific implementation, the target advertisement is delivered according to the second algorithm control parameters and the target advertisement delivery algorithm to improve the delivery effect of the target advertisement. In this way, the corresponding algorithm control parameters can be dynamically adjusted using advertisement feedback, thereby improving the delivery effect of the target advertisement.
[0115] It can be seen that the big data-based advertising delivery optimization method described in the embodiment of the present application obtains target attribute information of the target advertisement, uses big data technology to match the attribute information of Q advertisements in the advertisement database with the target attribute information, obtains P attribute information that successfully matches the target attribute information, obtains reference advertisements corresponding to the P attribute information, obtains P reference advertisements, each of the Q advertisements corresponds to one attribute information, P is an integer greater than 1, Q is an integer greater than or equal to P, obtains advertising delivery algorithms corresponding to the P reference advertisements, obtains M advertising delivery algorithms, M is a positive integer less than or equal to P, selects a target advertising delivery algorithm from the M advertising delivery algorithms, obtains a first algorithm control parameter corresponding to the target advertising delivery algorithm, and calculates the advertising delivery algorithm based on the first algorithm control parameter and The target advertisement delivery algorithm delivers the target advertisement, obtains the target feedback information for the target advertisement, optimizes the first algorithm control parameter according to the target feedback information, obtains the second algorithm control parameter, delivers the target advertisement according to the second algorithm control parameter and the target advertisement delivery algorithm, so as to improve the delivery effect of the target advertisement. In this way, big data can be used to select a target advertisement delivery algorithm that is adapted to the attributes of the target advertisement, and deliver the target advertisement with the first algorithm control parameter under the target advertisement delivery algorithm. The algorithm control parameters of the target advertisement delivery algorithm can also be adjusted based on the dynamic feedback of the delivery effect of the target advertisement, so that the optimized advertising effect is further improved and more in line with the current actual situation, so that the target advertisement delivery algorithm can deliver the target advertisement more adaptively, thereby improving the advertising publicity effect.
[0116] In accordance with the above embodiment, please refer to Figure 2 , Figure 2 : is a structural diagram of a server provided in an embodiment of the present application. As shown in the figure, the server includes a processor, a memory, a communication interface, and one or more programs. The one or more programs are stored in the memory and configured to be executed by the processor. In the embodiment of the present application, the program includes instructions for executing the following steps:
[0117] Obtain target attribute information of the target advertisement;
[0118] Using big data technology, the attribute information of Q advertisements in the advertisement database is matched with the target attribute information to obtain P attribute information that successfully matches the target attribute information, and reference advertisements corresponding to the P attribute information are obtained to obtain P reference advertisements, where each advertisement in the Q advertisements corresponds to one attribute information, where P is an integer greater than 1, and Q is an integer greater than or equal to P;
[0119] Obtaining advertisement delivery algorithms corresponding to the P reference advertisements to obtain M advertisement delivery algorithms, where M is a positive integer less than or equal to P;
[0120] Selecting a target advertisement delivery algorithm from the M advertisement delivery algorithms;
[0121] Obtaining a first algorithm control parameter corresponding to the target advertisement delivery algorithm;
[0122] Delivering the target advertisement according to the first algorithm control parameter and the target advertisement delivery algorithm;
[0123] Obtaining target feedback information for the target advertisement;
[0124] Optimize the first algorithm control parameter according to the target feedback information to obtain the second algorithm control parameter;
[0125] The target advertisement is delivered according to the second algorithm control parameter and the target advertisement delivery algorithm to improve the delivery effect of the target advertisement.
[0126] Optionally, in the aspect of selecting a target advertisement delivery algorithm from the M advertisement delivery algorithms, the program includes instructions for executing the following steps:
[0127] Obtaining push result information and algorithm evaluation parameters of each of the M advertisement delivery algorithms, and obtaining m push result information and m algorithm evaluation parameters;
[0128] Determine a first evaluation value corresponding to each piece of push result information in the m pieces of push result information to obtain m first evaluation values;
[0129] Determine a second evaluation value corresponding to each of the m algorithm evaluation parameters to obtain m second evaluation values;
[0130] Performing a weighted operation according to the m first evaluation values and the m second evaluation values to obtain m target evaluation values;
[0131] A maximum value among the m target evaluation values is selected, and the advertisement delivery algorithm corresponding to the maximum value is determined as the target advertisement delivery algorithm.
[0132] Optionally, in the aspect of determining the first evaluation value corresponding to each of the m pieces of push result information to obtain the m first evaluation values, the program includes instructions for executing the following steps:
[0133] Acquire first push result information, the first push result information including at least one advertisement push result information, each push result information including: the number of potential push objects, the number of actual push objects, and advertisement browsing status; the first push result information is any one of the m push result information; the advertisement browsing status includes at least one of the following: browsing time, browsing attention content, and content time period for browsing advertisement content;
[0134] Determine a first ratio between the actual number of push objects and the potential number of push objects in the push result information of each advertisement in the push result information of the at least one advertisement, and determine an advertisement browsing index factor according to the advertisement browsing situation;
[0135] Determine at least one evaluation value according to the first ratio of each advertisement in the at least one advertisement and the advertisement browsing index factor;
[0136] Acquire attribute information of the at least one advertisement to obtain at least one attribute information;
[0137] Determine a weight corresponding to each attribute information in the at least one attribute information to obtain at least one weight;
[0138] A weighted operation is performed on the at least one weight and the at least one evaluation value to obtain the first evaluation value.
[0139] Optionally, in determining at least one evaluation value according to the first ratio of each advertisement in the at least one advertisement and the advertisement browsing index factor, the program includes instructions for executing the following steps:
[0140] Obtaining a target first ratio and a target advertisement browsing index factor for a first advertisement, wherein the first advertisement is any advertisement among the at least one advertisement;
[0141] determining a target adjustment parameter corresponding to the target first ratio;
[0142] Determining a reference evaluation value corresponding to the target advertisement browsing index factor;
[0143] Obtaining a target first score for the first advertisement, where the target first score is related to a preset operation for the first advertisement, where the preset operation includes at least one of the following: like, favorite, comment, and forward;
[0144] determining a target fine-tuning parameter corresponding to the target first score;
[0145] The reference evaluation value is adjusted according to the target adjustment parameter and the target fine-tuning parameter to obtain the evaluation value of the first advertisement.
[0146] Optionally, the target feedback information includes target statistical information of designated operations for the target advertisement within a preset time period, and the designated operations include: like, favorite, comment, forward, download, blacklist, and dislike;
[0147] In terms of optimizing the first algorithm control parameter according to the target feedback information to obtain the second algorithm control parameter, the above program includes instructions for executing the following steps:
[0148] Determine the target advertisement popularity according to the target statistical information and the length of the preset time period;
[0149] Determine the positive evaluation result statistical information and the negative evaluation result statistical information of the target advertisement according to the target statistical information;
[0150] Determining a target ratio between the positive evaluation result statistical information and the negative evaluation result statistical information;
[0151] Obtaining a target reference ratio corresponding to the target advertisement popularity;
[0152] determining a target deviation value between the target ratio and the target reference ratio;
[0153] Determining a target feedback optimization coefficient corresponding to the target deviation;
[0154] The first algorithm control parameter is optimized according to the target feedback optimization coefficient to obtain the second algorithm control parameter.
[0155] Optionally, in the aspect of delivering the target advertisement according to the first algorithm control parameter and the target advertisement delivery algorithm, the program includes instructions for executing the following steps:
[0156] Determine k delivery objects according to the first algorithm control parameter and the target advertisement delivery algorithm, where k is a positive integer;
[0157] Obtaining habit parameters of a target user of a first delivery object; the first delivery object is any delivery object among the k delivery objects;
[0158] Determining target push parameters of the target advertisement according to the target user habit parameters;
[0159] The target advertisement is pushed to the first delivery object according to the target push parameter.
[0160] It can be seen that the server described in the embodiment of the present application obtains target attribute information of the target advertisement, uses big data technology to match the attribute information of Q advertisements in the advertisement database with the target attribute information, obtains P attribute information that successfully matches the target attribute information, obtains reference advertisements corresponding to the P attribute information, obtains P reference advertisements, each of the Q advertisements corresponds to one attribute information, P is an integer greater than 1, Q is an integer greater than or equal to P, obtains advertisement delivery algorithms corresponding to the P reference advertisements, obtains M advertisement delivery algorithms, M is a positive integer less than or equal to P, selects a target advertisement delivery algorithm from the M advertisement delivery algorithms, obtains a first algorithm control parameter corresponding to the target advertisement delivery algorithm, and selects the first algorithm control parameter according to the first algorithm control parameter and the target advertisement delivery algorithm. A target advertisement is delivered using a target advertisement delivery algorithm, target feedback information for the target advertisement is obtained, the first algorithm control parameter is optimized according to the target feedback information, the second algorithm control parameter is obtained, and the target advertisement is delivered according to the second algorithm control parameter and the target advertisement delivery algorithm to improve the delivery effect of the target advertisement. In this way, big data can be used to select a target advertisement delivery algorithm that is adapted to the target advertisement's own attributes, and the target advertisement is delivered using the first algorithm control parameter under the target advertisement delivery algorithm. The algorithm control parameters of the target advertisement delivery algorithm can also be adjusted based on dynamic feedback of the delivery effect of the target advertisement, so that the optimized advertising effect is further improved and more in line with the current actual situation, so that the target advertisement delivery algorithm can deliver the target advertisement more adaptively, thereby improving the advertising effect.
[0161] Figure 3 This is a functional unit block diagram of an advertising delivery optimization system 300 based on big data involved in the embodiment of the present application. The advertising delivery optimization system 300 based on big data includes: an acquisition unit 301, a selection unit 302, a delivery unit 303, and an optimization unit 304, wherein:
[0162] The acquisition unit 301 is used to acquire target attribute information of a target advertisement; use big data technology to match the attribute information of Q advertisements in an advertisement database with the target attribute information to obtain P attribute information that successfully matches the target attribute information, acquire reference advertisements corresponding to the P attribute information, and obtain P reference advertisements, where each of the Q advertisements corresponds to one attribute information, P is an integer greater than 1, and Q is an integer greater than or equal to P; acquire advertisement delivery algorithms corresponding to the P reference advertisements to obtain M advertisement delivery algorithms, where M is a positive integer less than or equal to P;
[0163] A selection unit 302, configured to select a target advertisement delivery algorithm from the M advertisement delivery algorithms;
[0164] The acquisition unit 301 is further used to acquire a first algorithm control parameter corresponding to the target advertisement delivery algorithm;
[0165] The delivery unit 303 is used to deliver the target advertisement according to the first algorithm control parameter and the target advertisement delivery algorithm;
[0166] The acquisition unit 301 is further configured to acquire target feedback information for the target advertisement;
[0167] The optimization unit 304 is used to optimize the first algorithm control parameter according to the target feedback information to obtain the second algorithm control parameter;
[0168] The delivery unit 303 is further configured to deliver the target advertisement according to the second algorithm control parameter and the target advertisement delivery algorithm to improve delivery effect of the target advertisement.
[0169] Optionally, in the aspect of selecting a target advertisement delivery algorithm from the M advertisement delivery algorithms, the selection unit 302 is specifically configured to:
[0170] Obtaining push result information and algorithm evaluation parameters of each of the M advertisement delivery algorithms, and obtaining m push result information and m algorithm evaluation parameters;
[0171] Determine a first evaluation value corresponding to each piece of push result information in the m pieces of push result information to obtain m first evaluation values;
[0172] Determine a second evaluation value corresponding to each of the m algorithm evaluation parameters to obtain m second evaluation values;
[0173] Performing a weighted operation according to the m first evaluation values and the m second evaluation values to obtain m target evaluation values;
[0174] A maximum value among the m target evaluation values is selected, and the advertisement delivery algorithm corresponding to the maximum value is determined as the target advertisement delivery algorithm.
[0175] Optionally, in determining the first evaluation value corresponding to each of the m pieces of push result information to obtain the m first evaluation values, the selection unit 302 is specifically configured to:
[0176] Acquire first push result information, the first push result information including at least one advertisement push result information, each push result information including: the number of potential push objects, the number of actual push objects, and advertisement browsing status; the first push result information is any one of the m push result information; the advertisement browsing status includes at least one of the following: browsing time, browsing attention content, and content time period for browsing advertisement content;
[0177] Determine a first ratio between the actual number of push objects and the potential number of push objects in the push result information of each advertisement in the push result information of the at least one advertisement, and determine an advertisement browsing index factor according to the advertisement browsing situation;
[0178] Determine at least one evaluation value according to the first ratio of each advertisement in the at least one advertisement and the advertisement browsing index factor;
[0179] Acquire attribute information of the at least one advertisement to obtain at least one attribute information;
[0180] Determine a weight corresponding to each attribute information in the at least one attribute information to obtain at least one weight;
[0181] A weighted operation is performed on the at least one weight and the at least one evaluation value to obtain the first evaluation value.
[0182] Optionally, in determining at least one evaluation value according to the first ratio of each advertisement in the at least one advertisement and the advertisement browsing index factor, the selection unit 302 is specifically configured to:
[0183] Obtaining a target first ratio and a target advertisement browsing index factor for a first advertisement, wherein the first advertisement is any advertisement among the at least one advertisement;
[0184] determining a target adjustment parameter corresponding to the target first ratio;
[0185] Determining a reference evaluation value corresponding to the target advertisement browsing index factor;
[0186] Obtaining a target first score for the first advertisement, where the target first score is related to a preset operation for the first advertisement, where the preset operation includes at least one of the following: like, favorite, comment, and forward;
[0187] determining a target fine-tuning parameter corresponding to the target first score;
[0188] The reference evaluation value is adjusted according to the target adjustment parameter and the target fine-tuning parameter to obtain the evaluation value of the first advertisement.
[0189] Optionally, the target feedback information includes target statistical information of designated operations for the target advertisement within a preset time period, and the designated operations include: like, favorite, comment, forward, download, blacklist, and dislike;
[0190] In terms of optimizing the first algorithm control parameter according to the target feedback information to obtain the second algorithm control parameter, the optimization unit 304 is specifically used to:
[0191] Determine the target advertisement popularity according to the target statistical information and the length of the preset time period;
[0192] Determine the positive evaluation result statistical information and the negative evaluation result statistical information of the target advertisement according to the target statistical information;
[0193] Determining a target ratio between the positive evaluation result statistical information and the negative evaluation result statistical information;
[0194] Obtaining a target reference ratio corresponding to the target advertisement popularity;
[0195] determining a target deviation value between the target ratio and the target reference ratio;
[0196] Determining a target feedback optimization coefficient corresponding to the target deviation;
[0197] The first algorithm control parameter is optimized according to the target feedback optimization coefficient to obtain the second algorithm control parameter.
[0198] Optionally, in the aspect of delivering the target advertisement according to the first algorithm control parameter and the target advertisement delivery algorithm, the delivery unit 303 is specifically configured to:
[0199] Determine k delivery objects according to the first algorithm control parameter and the target advertisement delivery algorithm, where k is a positive integer;
[0200] Obtaining habit parameters of a target user of a first delivery object; the first delivery object is any delivery object among the k delivery objects;
[0201] Determining target push parameters of the target advertisement according to the target user habit parameters;
[0202] The target advertisement is pushed to the first delivery object according to the target push parameter.
[0203] It can be seen that the big data-based advertising delivery optimization system described in the embodiment of the present application obtains target attribute information of the target advertisement, uses big data technology to match the attribute information of Q advertisements in the advertising database with the target attribute information, obtains P attribute information that successfully matches the target attribute information, obtains reference advertisements corresponding to the P attribute information, obtains P reference advertisements, each of the Q advertisements corresponds to one attribute information, P is an integer greater than 1, Q is an integer greater than or equal to P, obtains advertising delivery algorithms corresponding to the P reference advertisements, obtains M advertising delivery algorithms, M is a positive integer less than or equal to P, selects a target advertising delivery algorithm from the M advertising delivery algorithms, obtains a first algorithm control parameter corresponding to the target advertising delivery algorithm, and calculates the advertising delivery algorithm based on the first algorithm control parameter and The target advertisement delivery algorithm delivers the target advertisement, obtains the target feedback information for the target advertisement, optimizes the first algorithm control parameter according to the target feedback information, obtains the second algorithm control parameter, delivers the target advertisement according to the second algorithm control parameter and the target advertisement delivery algorithm, so as to improve the delivery effect of the target advertisement. In this way, big data can be used to select a target advertisement delivery algorithm that is adapted to the attributes of the target advertisement, and deliver the target advertisement with the first algorithm control parameter under the target advertisement delivery algorithm. The algorithm control parameters of the target advertisement delivery algorithm can also be adjusted based on the dynamic feedback of the delivery effect of the target advertisement, so that the optimized advertising effect is further improved and more in line with the current actual situation, so that the target advertisement delivery algorithm can deliver the target advertisement more adaptively, thereby improving the advertising publicity effect.
[0204] An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute part or all of the steps of any method recorded in the above method embodiments.
[0205] The present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps of any method described in the above method embodiment. The computer program product may be a software installation package.
[0206] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.
[0207] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0208] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.
[0209] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0210] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0211] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the above-mentioned methods in each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.
[0212] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, and the memory can include: a flash drive, a read-only memory (English: Read-Only Memory, abbreviated: ROM), a random access memory (English: Random Access Memory, abbreviated: RAM), a magnetic disk or an optical disk, etc.
[0213] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for general technical personnel in this field, according to the idea of the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.
Claims
1. A method for optimizing advertising delivery based on big data, characterized in that: The method comprises: Obtain target attribute information of the target advertisement; Using big data technology, the attribute information of Q advertisements in the advertisement database is matched with the target attribute information to obtain P attribute information that successfully matches the target attribute information, and reference advertisements corresponding to the P attribute information are obtained to obtain P reference advertisements, where each advertisement in the Q advertisements corresponds to one attribute information, where P is an integer greater than 1, and Q is an integer greater than or equal to P; Obtaining advertisement delivery algorithms corresponding to the P reference advertisements to obtain M advertisement delivery algorithms, where M is a positive integer less than or equal to P; Selecting a target advertisement delivery algorithm from the M advertisement delivery algorithms; Obtaining a first algorithm control parameter corresponding to the target advertisement delivery algorithm; Delivering the target advertisement according to the first algorithm control parameter and the target advertisement delivery algorithm; Obtaining target feedback information for the target advertisement; Optimize the first algorithm control parameter according to the target feedback information to obtain the second algorithm control parameter; delivering the target advertisement according to the second algorithm control parameter and the target advertisement delivery algorithm to improve the delivery effect of the target advertisement; The step of selecting a target advertisement delivery algorithm from the M advertisement delivery algorithms includes: Obtaining push result information and algorithm evaluation parameters of each of the M advertisement delivery algorithms, and obtaining m push result information and m algorithm evaluation parameters; Determine a first evaluation value corresponding to each piece of push result information in the m pieces of push result information to obtain m first evaluation values; Determine a second evaluation value corresponding to each of the m algorithm evaluation parameters to obtain m second evaluation values; Performing a weighted operation according to the m first evaluation values and the m second evaluation values to obtain m target evaluation values; Selecting a maximum value among the m target evaluation values, and determining the advertisement delivery algorithm corresponding to the maximum value as the target advertisement delivery algorithm; The determining of the first evaluation value corresponding to each of the m push result information to obtain the m first evaluation values includes: Acquire first push result information, the first push result information including at least one advertisement push result information, each push result information including: the number of potential push objects, the number of actual push objects, and advertisement browsing status; the first push result information is any one of the m push result information; the advertisement browsing status includes at least one of the following: browsing time, browsing attention content, and content time period of browsing advertisement content; Determine a first ratio between the actual number of push objects and the potential number of push objects in the push result information of each advertisement in the push result information of the at least one advertisement, and determine an advertisement browsing index factor according to the advertisement browsing situation; Determine at least one evaluation value according to the first ratio of each advertisement in the at least one advertisement and the advertisement browsing index factor; Acquire attribute information of the at least one advertisement to obtain at least one attribute information; Determine a weight corresponding to each attribute information in the at least one attribute information to obtain at least one weight; Performing a weighted operation on the at least one weight and the at least one evaluation value to obtain the first evaluation value; The step of determining at least one evaluation value according to the first ratio of each advertisement in the at least one advertisement and the advertisement browsing index factor comprises: Obtaining a target first ratio and a target advertisement browsing index factor for a first advertisement, wherein the first advertisement is any advertisement among the at least one advertisement; determining a target adjustment parameter corresponding to the target first ratio; Determining a reference evaluation value corresponding to the target advertisement browsing index factor; Obtaining a target first score for the first advertisement, where the target first score is related to a preset operation for the first advertisement, where the preset operation includes at least one of the following: like, favorite, comment, and forward; determining a target fine-tuning parameter corresponding to the target first score; The reference evaluation value is adjusted according to the target adjustment parameter and the target fine-tuning parameter to obtain the evaluation value of the first advertisement.
2. The method according to claim 1, characterized in that The target feedback information includes target statistical information of designated operations for the target advertisement within a preset time period, and the designated operations include: like, favorite, comment, forward, download, blacklist, and dislike; The step of optimizing the first algorithm control parameter according to the target feedback information to obtain the second algorithm control parameter includes: Determine the target advertisement popularity according to the target statistical information and the length of the preset time period; Determine the positive evaluation result statistical information and the negative evaluation result statistical information of the target advertisement according to the target statistical information; Determining a target ratio between the positive evaluation result statistical information and the negative evaluation result statistical information; Obtaining a target reference ratio corresponding to the target advertisement popularity; determining a target deviation value between the target ratio and the target reference ratio; Determining a target feedback optimization coefficient corresponding to the target deviation value; The first algorithm control parameter is optimized according to the target feedback optimization coefficient to obtain the second algorithm control parameter.
3. The method according to claim 1, characterized in that: The delivering the target advertisement according to the first algorithm control parameter and the target advertisement delivery algorithm comprises: Determine k delivery objects according to the first algorithm control parameter and the target advertisement delivery algorithm, where k is a positive integer; Obtaining habit parameters of a target user of a first delivery object; the first delivery object is any delivery object among the k delivery objects; Determining target push parameters of the target advertisement according to the target user habit parameters; The target advertisement is pushed to the first delivery object according to the target push parameter.
4. An advertising delivery optimization system based on big data, characterized in that: The system includes: an acquisition unit, a selection unit, a delivery unit, and an optimization unit, wherein: The acquisition unit is used to acquire target attribute information of a target advertisement; use big data technology to match attribute information of Q advertisements in an advertisement database with the target attribute information to obtain P attribute information that successfully matches the target attribute information, acquire reference advertisements corresponding to the P attribute information, and obtain P reference advertisements, where each of the Q advertisements corresponds to one attribute information, P is an integer greater than 1, and Q is an integer greater than or equal to P; acquire advertisement delivery algorithms corresponding to the P reference advertisements to obtain M advertisement delivery algorithms, where M is a positive integer less than or equal to P; A selection unit, configured to select a target advertisement delivery algorithm from the M advertisement delivery algorithms; The acquisition unit is further used to acquire a first algorithm control parameter corresponding to the target advertisement delivery algorithm; The delivery unit is configured to deliver the target advertisement according to the first algorithm control parameter and the target advertisement delivery algorithm; The acquisition unit is further used to acquire target feedback information for the target advertisement; The optimization unit is used to optimize the first algorithm control parameter according to the target feedback information to obtain the second algorithm control parameter; The delivery unit is further configured to deliver the target advertisement according to the second algorithm control parameter and the target advertisement delivery algorithm to improve the delivery effect of the target advertisement; Wherein, in the aspect of selecting the target advertisement delivery algorithm from the M advertisement delivery algorithms, the selection unit is specifically used for: Obtaining push result information and algorithm evaluation parameters of each of the M advertisement delivery algorithms, and obtaining m push result information and m algorithm evaluation parameters; Determine a first evaluation value corresponding to each piece of push result information in the m pieces of push result information to obtain m first evaluation values; Determine a second evaluation value corresponding to each of the m algorithm evaluation parameters to obtain m second evaluation values; Performing a weighted operation according to the m first evaluation values and the m second evaluation values to obtain m target evaluation values; Selecting a maximum value among the m target evaluation values, and determining the advertisement delivery algorithm corresponding to the maximum value as the target advertisement delivery algorithm; Wherein, in determining the first evaluation value corresponding to each of the m pieces of push result information to obtain the m first evaluation values, the selection unit is specifically used to: Acquire first push result information, the first push result information including at least one advertisement push result information, each push result information including: the number of potential push objects, the number of actual push objects, and advertisement browsing status; the first push result information is any one of the m push result information; the advertisement browsing status includes at least one of the following: browsing time, browsing attention content, and content time period of browsing advertisement content; Determine a first ratio between the actual number of push objects and the potential number of push objects in the push result information of each advertisement in the push result information of the at least one advertisement, and determine an advertisement browsing index factor according to the advertisement browsing situation; Determine at least one evaluation value according to the first ratio of each advertisement in the at least one advertisement and the advertisement browsing index factor; Acquire attribute information of the at least one advertisement to obtain at least one attribute information; Determine a weight corresponding to each attribute information in the at least one attribute information to obtain at least one weight; Performing a weighted operation on the at least one weight and the at least one evaluation value to obtain the first evaluation value; Wherein, in determining at least one evaluation value according to the first ratio of each advertisement in the at least one advertisement and the advertisement browsing index factor, the selection unit is specifically used to: Obtaining a target first ratio and a target advertisement browsing index factor for a first advertisement, wherein the first advertisement is any advertisement among the at least one advertisement; determining a target adjustment parameter corresponding to the target first ratio; Determining a reference evaluation value corresponding to the target advertisement browsing index factor; Obtaining a target first score for the first advertisement, where the target first score is related to a preset operation for the first advertisement, where the preset operation includes at least one of the following: like, favorite, comment, and forward; determining a target fine-tuning parameter corresponding to the target first score; The reference evaluation value is adjusted according to the target adjustment parameter and the target fine-tuning parameter to obtain the evaluation value of the first advertisement.
5. The system according to claim 4, characterized in that The target feedback information includes target statistical information of designated operations for the target advertisement within a preset time period, and the designated operations include: like, favorite, comment, forward, download, blacklist, and dislike; The step of optimizing the first algorithm control parameter according to the target feedback information to obtain the second algorithm control parameter includes: Determine the target advertisement popularity according to the target statistical information and the length of the preset time period; Determine the positive evaluation result statistical information and the negative evaluation result statistical information of the target advertisement according to the target statistical information; Determining a target ratio between the positive evaluation result statistical information and the negative evaluation result statistical information; Obtaining a target reference ratio corresponding to the target advertisement popularity; determining a target deviation value between the target ratio and the target reference ratio; Determining a target feedback optimization coefficient corresponding to the target deviation value; The first algorithm control parameter is optimized according to the target feedback optimization coefficient to obtain the second algorithm control parameter.
