Advertisement putting strategy generation method and terminal equipment
By combined with advertising media portraits and release indicators, the program recommendation model is used to generate an advertising strategy, which solves the problem of inefficient efficiency when choosing an advertising media, and achieves more accurate advertising and higher conversion rates.
Patent Information
- Application Number
- CN202411424168.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-12
- Publication Date
- 2025-01-24
AI Technical Summary
When faced with diverse advertising media choices, it is difficult for advertisers to make optimal decisions in a short period of time, resulting in inefficient advertising strategy generation.
By obtaining advertising media portrait data and entering the media portrait model, we can obtain advertising indicators and media portraits. The input scheme recommended model is used to generate a recommended launch strategy. This method combines the types of media, audience portraits, and the weight of the launching indicators, and accurately match advertising media and target audiences.
It improves the generation efficiency of advertising strategies, improves the accuracy of advertising positioning and conversion rate, ensures that advertising content is highly consistent with target users, and maximizes advertising effects.
Smart Images
Figure CN119359374A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of advertising delivery, and in particular to an advertising delivery strategy generation method and a terminal device. Background Art
[0002] There are a large number of advertising media in the current market, including but not limited to websites, mobile applications, social media platforms, video platforms, search engines, etc. Each media has its own unique user groups, traffic characteristics and advertising forms. Advertisers need to choose the most appropriate media for delivery based on their own advertising goals and the characteristics of the target audience. However, faced with such a variety of choices, advertisers often find it difficult to make the best decision in a short period of time.
[0003] Therefore, how to quickly generate advertising delivery strategies has become an urgent problem that needs to be solved. Summary of the invention
[0004] The present application provides an advertisement delivery strategy generation and terminal device, which can improve the efficiency of generating advertisement delivery strategies.
[0005] In a first aspect, the present application provides a method for generating an advertisement delivery strategy, comprising:
[0006] Acquire multiple advertising media portrait data, input the multiple advertising media portrait data into the media portrait model, and obtain multiple advertising media portraits, the multiple advertising media portraits include advertising media types and advertising media audience portraits; obtain multiple advertising delivery indicators, each of the multiple advertising delivery indicators has a corresponding delivery indicator weight; input the multiple advertising delivery indicators and the multiple advertising media portraits into the solution recommendation model, and obtain multiple recommended delivery strategies.
[0007] By adopting the above technical solution, by utilizing the advertising media type and high conversion rate audience information analyzed by the media portrait model, combined with various delivery indicators and their weights, it is possible to accurately match advertising media with target audiences, which not only improves the accuracy of advertising positioning, but also enhances the fit between advertising content and target users. This improves the efficiency of generating advertising delivery strategies while ensuring the accuracy and conversion rate of advertising delivery.
[0008] Optionally, after inputting multiple advertising delivery indicators and multiple advertising media portraits into the solution recommendation model to obtain multiple recommended delivery strategies, the method also includes: inputting multiple advertising delivery indicators and multiple advertising media portraits into the solution recommendation model to obtain multiple recommended delivery strategies; obtaining first media portraits and target audience portraits corresponding to the multiple recommended delivery strategies; calculating the initial similarity value of the first media portrait and the target audience portrait through preset portrait parameter weights and a preset similarity algorithm; and sorting the multiple recommended delivery strategies according to the initial similarity value to obtain a list of recommended delivery strategies.
[0009] By adopting the above technical solution and calculating the initial similarity value, we can ensure that the recommended media is highly matched with the target audience. Advertisers can choose the most appropriate media combination more scientifically, further improve the accuracy and effectiveness of advertising, and maximize the effect of advertising.
[0010] Optionally, before sorting multiple recommended delivery strategies according to the initial similarity value to obtain a list of recommended delivery strategies, the method also includes: obtaining a product delivery record corresponding to each of the multiple recommended delivery strategies; calculating, based on the product delivery record and multiple advertising delivery indicators, a second similarity value of the product information in the product delivery record and the multiple advertising delivery indicators; determining whether the second similarity value is greater than a preset product similarity value; if greater, calculating the sum of the second similarity value and the initial similarity value based on the correspondence between the second similarity value and the initial similarity value, sorting the multiple recommended delivery strategies according to the calculation results, and obtaining a list of recommended delivery strategies.
[0011] By adopting the above technical solution, by introducing product delivery records and second similarity values, the recommendation strategy can be further refined to ensure that the recommended media is not only highly matched with the target audience, but also consistent with past successful delivery experience, thereby further improving the accuracy and effectiveness of advertising.
[0012] Optionally, after calculating the sum of the second similarity value and the initial similarity value according to the correspondence between the second similarity value and the initial similarity value, sorting multiple recommended delivery strategies according to the calculation result, and obtaining a recommended delivery strategy list, the method also includes: determining whether there are multiple online recommended delivery strategies among the multiple recommended delivery strategies; if so, obtaining the audience historical behavior corresponding to the second similarity value of the multiple online recommended delivery strategies being greater than a preset product similarity value; sorting the multiple recommended delivery strategies according to the audience historical behavior and multiple advertising delivery indicators, and generating a behavior recommendation list.
[0013] By adopting the above technical solutions and introducing the audience's historical behavior data, multiple recommendation delivery strategies can be further optimized to ensure that the recommended media is not only consistent with the target audience and past successful delivery experience, but can also be fine-tuned based on actual historical behavior performance, thereby maximizing advertising effects and improving conversion rates and return on investment.
[0014] Optionally, after calculating the initial similarity value between the first media portrait and the target audience portrait by using preset portrait parameter weights and a preset similarity algorithm, the method also includes: determining whether the initial similarity value is greater than the standard similarity value; if greater, generating corresponding recommended advertising features for each advertising media in multiple recommended delivery strategies based on the first media portrait, so as to generate multiple recommended delivery strategies based on the advertising recommendation features.
[0015] By adopting the above technical solution and generating recommended advertising features, it is ensured that the recommended advertising content is highly matched with the audience characteristics of the media, further improving the accuracy and attractiveness of the advertisements, increasing user click-through rates and conversion rates, and thus optimizing the advertising effect.
[0016] Optionally, before acquiring multiple advertising media portrait data and inputting the multiple advertising media portrait data into a media portrait model to obtain multiple advertising media portraits, the method also includes: acquiring media portrait data, the media portrait data including media advertising unit price, historical delivery effects, and audience portrait; processing the media portrait data through a feature conversion method to obtain training set data, the feature conversion method including one-hot encoding; training the initial machine learning model through the training data set data to obtain a media portrait model.
[0017] By adopting the above technical solution, the initial machine learning model is trained with media portrait data, so that the media portrait model can more accurately identify the characteristics of the media and the target audience, optimize the advertising delivery strategy, and improve the advertising effect and return on investment.
[0018] Optionally, before inputting multiple advertising delivery indicators and multiple advertising media portraits into the solution recommendation model to obtain multiple recommended delivery strategies, the method also includes: obtaining multiple historical advertising delivery indicators and multiple advertising media portraits; processing the multiple historical advertising delivery indicators and multiple advertising media portraits through feature engineering to obtain a feature data set, the feature engineering including multi-table feature concatenation and wide table feature combination; training the initial regression model with the feature data set to generate a solution recommendation model.
[0019] By adopting the above technical solutions, the generated solution recommendation model can comprehensively consider multiple advertising indicators and media portraits, accurately recommend the most suitable media combination, optimize advertising effects, increase brand exposure and sales conversion rate, and reduce advertising costs.
[0020] In the second aspect of the present application, a terminal device is provided, including: a media portrait acquisition module, used to acquire multiple advertising media portrait data, input the multiple advertising media portrait data into a media portrait model, and obtain multiple advertising media portraits, the multiple advertising media portraits include the type of advertising media and the advertising media affected by multiple advertising delivery indicators and multiple advertising media portrait portraits; a delivery indicator acquisition module, used to acquire multiple advertising delivery indicators, each of the multiple advertising delivery indicators has a corresponding delivery indicator weight; a strategy acquisition module, used to input a plan recommendation model to acquire multiple recommended delivery strategies.
[0021] In a third aspect of the present application, an electronic device is provided, comprising a processor (301), a memory (305), a user interface (303) and a network interface (304), wherein the memory (305) is used to store instructions, the user interface (303) and the network interface (304) are used to communicate with other devices, and the processor (301) is used to execute the instructions stored in the memory (305) so that the electronic device (300) executes any method as in the first aspect.
[0022] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, the method steps of any one of the first aspects are executed. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 It is a flow chart of the method for generating an advertisement delivery strategy disclosed in an embodiment of the present application.
[0024] Figure 2 It is a structural diagram of a terminal device disclosed in an embodiment of the present application.
[0025] Figure 3 It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application. DETAILED DESCRIPTION
[0026] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.
[0027] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.
[0028] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.
[0029] The following, combined Figure 1 A method for generating an advertisement delivery strategy in an embodiment of the present application is described in detail.
[0030] Step S101: The terminal device obtains a plurality of advertising media portrait data, inputs the plurality of advertising media portrait data into a media portrait model, and obtains a plurality of advertising media portraits, wherein the plurality of advertising media portraits include advertising media types and advertising media audience portraits.
[0031] Multiple advertising media portrait data refers to multi-dimensional data collected from different advertising media on advertising unit prices, historical delivery effects, audience characteristics, etc.
[0032] The type of advertising media refers to the category of advertising media, such as video websites, social media platforms, professional forums, etc. Advertising media audience portraits are used to describe the characteristics of the target audience of advertising media, including age, gender, region, interests, etc.
[0033] Multiple advertising media portraits are comprehensive information describing the characteristics of each media and audience characteristics generated by the media portrait model based on the advertising media portrait data. For example, the type of media can be "video, humor" or "forum, artificial intelligence, education" or other.
[0034] Exemplarily, multiple advertising media portrait data are as follows: Media A: video website, advertising unit price 10 yuan / 1,000 impressions, historical CTR 1.5%, CVR 0.5%, audience portrait (20-35 years old, 70% male, interest tags: technology, programming); Media B: social media platform, advertising unit price 8 yuan / 1,000 impressions, historical CTR 2.0%, CVR 0.3%, audience portrait (18-30 years old, 60% female, interest tags: fashion, technology); Media C: professional forum, advertising unit price 5 yuan / 1,000 impressions, historical CTR 1.0%, CVR 1.0%, audience portrait (25-40 years old, 80% male, interest tags: programming, technology); Media D: technology blog, advertising unit price 6 yuan / 1,000 impressions, historical CTR 1.2%, CVR 0.8%, audience portrait (20-35 years old, 75% male, interest tags: technology, programming); Media E: entertainment news website, advertising unit price 4 yuan / 1,000 impressions, historical CTR 0.8%, CVR 0.1%, audience portrait (18-45 years old, 50% male, interest tags: entertainment, sports).
[0035] It should be understood that when obtaining advertising media data, the methods for obtaining advertising media portrait data for different advertising media are different. For example, when obtaining the opening screen advertising data of an APP, the exposure rate, click-through rate, conversion rate, and purchase rate can be obtained; when obtaining display advertising data, the number of views of the display position, browsing time and other data can be obtained.
[0036] Optionally, before acquiring multiple advertising media portrait data and inputting the multiple advertising media portrait data into a media portrait model to obtain multiple advertising media portraits, the method also includes: acquiring media portrait data, the media portrait data including media advertising unit price, historical delivery effects, and audience portrait; processing the media portrait data through a feature conversion method to obtain training set data, the feature conversion method including one-hot encoding; training the initial machine learning model through the training data set data to obtain a media portrait model.
[0037] In combination with the above example, after obtaining the media portrait data corresponding to media A, media B, media C, media D, and media E respectively, the categorical data (such as gender, interest tags) are converted into numerical data through one-hot encoding, such as gender: (male, 1), (female, 0), interest tags: (technology, [1,0,0]), (programming, [0,1,0]), (fashion, [0,0,1]) to obtain a training data set, and the converted data set is used to train the initial machine learning model to generate a media portrait model, where the initial machine learning model can select models such as logistic regression, random forest, and support vector machine. The data set is then divided into a training set and a test set, with a ratio of 80% training set and 20% test set. The model is trained using the training set data, and the performance of the model is evaluated using the test set data. The trained initial machine learning model is evaluated by accuracy, recall rate, F1 score, etc. until the initial machine learning model meets the evaluation criteria to obtain a media portrait model.
[0038] In this embodiment, the initial machine learning model is trained by media portrait data, so that the media portrait model can more accurately identify the characteristics of the media and the target audience, optimize the advertising delivery strategy, and improve the advertising effect and return on investment.
[0039] Step S102: the terminal device obtains a plurality of advertisement delivery indicators, each of which has a corresponding delivery indicator weight.
[0040] Multiple advertising delivery indicators are standards set by advertisers to measure advertising effectiveness, such as brand exposure, user acquisition, sales conversion, etc.
[0041] For example, take the advertiser as an online education company, the advertising product is a newly launched programming course, the target audience is technology enthusiasts aged 20-35, and the advertising goals are brand exposure, user acquisition, and sales conversion. The weights of each indicator are 0.3, 0.2, and 0.5 respectively, and the budget is 100,000 yuan.
[0042] Step S103: The terminal device inputs multiple advertising delivery indicators and multiple advertising media portraits into the solution recommendation model to obtain multiple recommended delivery strategies.
[0043] In the present application, the generated multiple recommended delivery strategies may also be used to represent one or more recommended delivery strategies, and each strategy may include one advertising medium or multiple advertising media, which is not specifically limited in the present application.
[0044] Exemplarily, the solution recommendation model calculates the comprehensive scores of multiple advertising media portraits according to multiple recommended delivery strategies, determines multiple recommended delivery strategies according to the calculated scores, and obtains multiple recommended delivery strategies as two recommended delivery strategies. One recommended strategy is media E, that is, media E meets all multiple advertising delivery indicators, and the multiple recommended delivery strategies are: [Media E (95)]; the second recommended strategy is media A, media B, media C, media D, that is, media A, media B, media C, media D, that is, after combining these four media, multiple advertising delivery indicators can be achieved. Optionally, the multiple recommended delivery strategies displayed to the user by the terminal device are: [Strategy 1: Media E (95)], [Strategy 2: Media A (85), Media B (70), Media C (90), Media D (88)]. Multiple recommended delivery strategies can also be one recommended delivery strategy, such as: [Media A (85), Media B (70), Media C (90), Media D (88)].
[0045] The advertising delivery strategy generation method provided in the embodiment of the present application can accurately match advertising media with target audiences by utilizing the advertising media type and high conversion rate audience information analyzed by the media portrait model, combined with various delivery indicators and their weights. This not only improves the accuracy of advertising positioning, but also enhances the fit between advertising content and target users, thereby improving the efficiency of generating advertising delivery strategies while ensuring the accuracy and conversion rate of advertising delivery.
[0046] Optionally, before inputting multiple advertising delivery indicators and multiple advertising media portraits into the solution recommendation model to obtain multiple recommended delivery strategies, the method also includes: obtaining multiple historical advertising delivery indicators and multiple advertising media portraits; processing the multiple historical advertising delivery indicators and multiple advertising media portraits through feature engineering to obtain a feature data set, the feature engineering including multi-table feature concatenation and wide table feature combination; training the initial regression model with the feature data set to generate a solution recommendation model.
[0047] Exemplarily, multiple advertising delivery indicators are obtained: brand exposure (weight 0.3), user acquisition (weight 0.2), sales conversion (weight 0.5), advertising media portrait: comprehensive information of media A, media B, media C, media D, and media E; multiple advertising delivery indicators and multiple advertising media portrait data are merged into a complete feature table, and then wide table feature combination is performed to create new feature combinations, such as the product of advertising unit price and CTR, and the combination of CVR and audience matching. Multiple advertising delivery indicators and media portrait data are merged to generate a wide table containing all relevant information, and new features are created, such as advertising unit price × historical CTR, CVR × audience matching, etc. The data set processed by feature engineering is used as the feature data set, the initial regression model is trained, and a solution recommendation model is generated.
[0048] In this implementation, the generated solution recommendation model can comprehensively consider multiple advertising indicators and media portraits, accurately recommend the most appropriate media combination, optimize advertising effects, increase brand exposure and sales conversion rate, and reduce advertising costs.
[0049] In an optional embodiment, after inputting multiple advertising delivery indicators and multiple advertising media portraits into a solution recommendation model to obtain multiple recommended delivery strategies, the method also includes: inputting multiple advertising delivery indicators and multiple advertising media portraits into the solution recommendation model to obtain multiple recommended delivery strategies; obtaining first media portraits and target audience portraits corresponding to the multiple recommended delivery strategies; calculating the initial similarity value of the first media portrait and the target audience portrait through preset portrait parameter weights and a preset similarity algorithm; and sorting the multiple recommended delivery strategies according to the initial similarity value to obtain a list of recommended delivery strategies.
[0050] Combined with the above example, multiple recommended delivery strategies are one delivery strategy: [Media A (85), Media B (70), Media C (90), Media D (88)]. Among them, the scheme recommendation model scores each advertising media as follows: Media A: comprehensive score 85, Media B: comprehensive score 70, Media C: comprehensive score 90, Media D: comprehensive score 88, among which Media E has a comprehensive score of 50 and is not used as a media for advertising; Media A: audience portrait (20-35 years old, male 70%, interest tags: technology, programming), Media B: audience portrait (18-30 years old, female 60%, interest tags: fashion, technology) The first media portrait is Media C: audience portrait (2 5-40 years old, 80% male, interest tags: programming, technology), media D: audience portrait (20-35 years old, 75% male, interest tags: technology, programming); the target audience portrait is age: 20-35 years old, gender: mainly male, interest tags: technology, programming; the preset portrait parameter weights are age: 0.4, gender: 0.3, interest tags: 0.3, and the preset similarity algorithm is the cosine similarity algorithm; the initial similarity value calculated is media D: age similarity: 1.0 (20-35 years old vs 20-35 years old), gender similarity: 0.75 (male 75% vs mainly male), interest tag similarity: 0.9 (technology, programming vs technology, programming), comprehensive similarity: 0.4*1.0+0.3*0.75+0.3*0.9=0.905, media A: age similarity: 1.0 (20-35 years old vs Media C: Age similarity: 0.8 (25-40 years old vs 20-35 years old), gender similarity: 0.8 (80% male vs mainly male), interest tag similarity: 0.9 (technology, programming vs technology, programming), comprehensive similarity: 0.4*1.0+0.3*0.7+0.3*0.9=0.88, Media B: Age similarity: 0.6 (18-30 years old vs 20-35 years old), gender similarity: 0.4 (60% female vs. mainly male), interest tag similarity: 0.6 (fashion, technology vs. technology, programming), comprehensive similarity: 0.4*0.6+0.3*0.4+0.3*0.6=0.54, and the recommended delivery strategy list is Media D: 0.905, Media A: 0.88, Media C: 0.83, Media B: 0.54.
[0051] If multiple recommended delivery strategies are two recommended delivery strategies, one of which is media E, and the comprehensive similarity of media E is 0.955, then the recommended delivery strategy list obtained is: [Strategy 1: Media E (0.955)], [Strategy 2: Media D (0.905), Media A (0.88), Media C (0.83), Media B (0.54)].
[0052] In this implementation, by calculating the initial similarity value, it is ensured that the recommended media is highly matched with the target audience, and advertisers can more scientifically select the most appropriate media combination, further improve the accuracy and effectiveness of advertising, and maximize the effect of advertising.
[0053] In an optional embodiment, before sorting multiple recommended delivery strategies according to the initial similarity value to obtain a list of recommended delivery strategies, the method also includes: obtaining a product delivery record corresponding to each of the multiple recommended delivery strategies; calculating, based on the product delivery record and multiple advertising delivery indicators, a second similarity value of the product information in the product delivery record and the multiple advertising delivery indicators; determining whether the second similarity value is greater than a preset product similarity value; if greater, calculating the sum of the second similarity value and the initial similarity value based on the correspondence between the second similarity value and the initial similarity value, sorting the multiple recommended delivery strategies according to the calculation result, and obtaining a list of recommended delivery strategies.
[0054] Combined with the above example, the product delivery records obtained are: Media A: Programming courses were delivered in the past, click-through rate 1.4%, conversion rate 0.7%, Media B: Programming courses were delivered in the past, click-through rate 1.1%, conversion rate 0.5%, Media C: Programming courses were delivered in the past, click-through rate 1.2%, conversion rate 0.8%, Media D: Programming courses were delivered in the past, click-through rate 1.3%, conversion rate 0.9%; the preset product similarity value is 0.7, and the similarity calculation is performed to obtain To: Media A: Click rate similarity: 1.4% / 1.5%=0.93, conversion rate similarity: 0.7% / 0.5%=1.4, comprehensive similarity: (0.93+1.4) / 2=1.165, Media B: Click rate similarity: 1.1% / 1.5%=0.73, conversion rate similarity: 0.5% / 0.5%=1.0, comprehensive similarity: (0.73+1.0) / 2=0.865, Media C: Click rate similarity: 1.2% / 1.5%=0.8, conversion rate similarity: 0.8% / 0.5%=1.6, comprehensive similarity: (0.8+1.6) / 2=1.2, Media D: Click rate similarity: 1.3% / 1.5%=0.87, conversion rate similarity: 0.9% / 0.5%=1.8, comprehensive similarity: (0.87+1.8) / 2=1.335; that is, the second similarity values are all greater than the preset product similarity values, and the second similarity values are calculated to be similar to the initial similarity values. The sum of similar values gives: Media D: comprehensive score = 0.905 + 1.335 = 2.24, Media A: comprehensive score = 0.88 + 1.165 = 2.045, Media C: comprehensive score = 0.83 + 1.2 = 2.03, Media B: comprehensive score = 0.54 + 0.865 = 1.405. The recommended delivery strategy list is Media D: 2.24, Media A: 2.045, Media C: 2.03, Media B: 1.405.
[0055] In this implementation, by introducing product placement records and a second similarity value, the recommendation strategy is further refined to ensure that the recommended media is not only highly matched with the target audience, but also consistent with past successful placement experience, thereby further improving the accuracy and effectiveness of the advertisement.
[0056] In an optional implementation, after calculating the sum of the second similarity value and the initial similarity value based on the correspondence between the second similarity value and the initial similarity value, sorting multiple recommended delivery strategies according to the calculation result, and obtaining a recommended delivery strategy list, the method also includes: determining whether there are multiple online recommended delivery strategies among the multiple recommended delivery strategies; if so, obtaining the audience historical behavior corresponding to the second similarity value of the multiple online recommended delivery strategies being greater than the preset product similarity value; sorting the multiple recommended delivery strategies according to the audience historical behavior and multiple advertising delivery indicators, and generating a behavior recommendation list.
[0057] Combined with the above example, multiple recommended delivery strategies are Media D: 2.24, Media A: 2.045, Media C: 2.03, Media B: 1.405, and all of them are online multiple recommended delivery strategies. The obtained audience historical behaviors are Media D: audience historical behavior (click-through rate 1.3%, conversion rate 0.9%), Media A: audience historical behavior (click-through rate 1.4%, conversion rate 0.7%), Media C: audience historical behavior (click-through rate 1.2%, conversion rate 0.8%), Media B: audience historical behavior (click-through rate 1.1%, conversion rate 0.5%); the audience historical behavior weights set are click-through rate: 0.4, conversion rate: 0.6, and the comprehensive behavior score is calculated: Media D: comprehensive behavior score = 0.4×1.3%+0.6×0.9%=0.0052+0.0054=0.0106, Media A: comprehensive behavior score = 0.4×1.4%+0.6 × 0.7% = 0.0056 + 0.0042 = 0.0098, Media C: comprehensive behavior score = 0.4 × 1.2% + 0.6 × 0.8% = 0.0048 + 0.0048 = 0.0096, Media B: comprehensive behavior score = 0.4 × 1.1% + 0.6 × 0.5% = 0.0044 + 0.0030 = 0.0074; Final ranking: Media D: comprehensive score = 2.24 + 0.01 06=2.2506, Media A: comprehensive score=2.045+0.0098=2.0548, Media C: comprehensive score=2.03+0.0096=2.0396, Media B: comprehensive score=1.405+0.0074=1.4124, and the recommended delivery strategy list is Media D: 2.2506, Media A: 2.0548, Media C: 2.0396, Media B: 1.4124.
[0058] In this implementation, by introducing audience historical behavior data, multiple recommendation delivery strategies are further optimized to ensure that the recommended media is not only consistent with the target audience and past successful delivery experience, but can also be fine-tuned based on actual historical behavior performance, thereby maximizing advertising effects and improving conversion rates and return on investment.
[0059] In an optional embodiment, after calculating the initial similarity value between the first media portrait and the target audience portrait by using preset portrait parameter weights and a preset similarity algorithm, the method further includes: determining whether the initial similarity value is greater than the standard similarity value; if greater, generating corresponding recommended advertising features for each advertising media in multiple recommended delivery strategies based on the first media portrait, so as to generate multiple recommended delivery strategies based on the advertising recommendation features.
[0060] Advertisement recommendation features are used to represent advertisement features that are applicable to advertisement media. For example, the advertisement features of media A are professionalism and cutting-edge technology.
[0061] Combined with the above example, the calculated initial similarity values are media A: 0.88, media B: 0.54, media C: 0.83, media D: 0.905, the standard similarity value is 0.6, and the judgment results are media A: 0.88>0.6, media B: 0.54<0.6, media C: 0.83>0.6, media D: 0.905>0.6. The generated recommended advertising features are media A: specialization, cutting-edge technology, media C: specialization, technical depth, media D: cutting-edge technology, practical tutorials. The recommended delivery strategy list is media D: 0.905, recommended advertising features (cutting-edge technology, practical tutorials), media A: 0.88, recommended advertising features (specialization, cutting-edge technology), media C: 0.83, recommended advertising features (specialization, technical depth), media B: 0.54.
[0062] In this embodiment, by generating recommended advertising features, it is ensured that the recommended advertising content is highly matched with the audience characteristics of the media, further improving the accuracy and attractiveness of the advertisement, increasing user click-through rate and conversion rate, and thus optimizing the advertising effect.
[0063] It is understandable that the terminal device is required to achieve Figure 1 The functions shown include hardware and / or software modules that perform the corresponding functions. In combination with the steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application in combination with the embodiments, but such implementation should not be considered to be beyond the scope of the present application.
[0064] In this embodiment, the terminal device can be divided into functional modules according to the above method example. For example, different functional modules can be divided according to different functions, or two or more functions can be integrated into one processing module. The above integrated modules can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic and is only a logical function division. There may be other division methods in actual implementation.
[0065] In the case of dividing each functional module into corresponding functional modules, Figure 2A possible schematic diagram of the terminal device 200 involved in the above embodiment is shown, and the terminal device 200 includes: a media portrait acquisition module 201, used to acquire multiple advertising media portrait data, input the multiple advertising media portrait data into a media portrait model, and obtain multiple advertising media portraits, the multiple advertising media portraits include the type of advertising media and the advertising media audience portrait; a delivery indicator acquisition module 202, used to acquire multiple advertising delivery indicators, each of the multiple advertising delivery indicators has a corresponding delivery indicator weight; a strategy acquisition module 203, used to input multiple advertising delivery indicators and multiple advertising media portraits into a solution recommendation model, and obtain multiple recommended delivery strategies.
[0066] In an optional implementation method of the embodiment of the present application, the strategy acquisition module 203 is also used to input multiple advertising delivery indicators and multiple advertising media portraits into the solution recommendation model to obtain multiple recommended delivery strategies; obtain the first media portrait and the target audience portrait corresponding to the multiple recommended delivery strategies; calculate the initial similarity value of the first media portrait and the target audience portrait through preset portrait parameter weights and a preset similarity algorithm; sort the multiple recommended delivery strategies according to the initial similarity value to obtain a list of recommended delivery strategies.
[0067] In an optional implementation of the embodiment of the present application, the strategy acquisition module 203 is also used to obtain a product delivery record corresponding to each of the multiple recommended delivery strategies; calculate a second similarity value of the product delivered in the product delivery record and the product information in the multiple advertising delivery indicators based on the product delivery record and the multiple advertising delivery indicators; determine whether the second similarity value is greater than a preset product similarity value; if greater, calculate the sum of the second similarity value and the initial similarity value based on the correspondence between the second similarity value and the initial similarity value, sort the multiple recommended delivery strategies according to the calculation results, and obtain a recommended delivery strategy list.
[0068] In an optional implementation method of the embodiment of the present application, the strategy acquisition module 203 is also used to determine whether there are multiple online recommended delivery strategies among the multiple recommended delivery strategies; if so, obtain the audience historical behavior corresponding to the second similarity value of the multiple online recommended delivery strategies that is greater than the preset product similarity value; sort the multiple recommended delivery strategies according to the audience historical behavior and multiple advertising delivery indicators, and generate a behavior recommendation list.
[0069] In an optional implementation method of the embodiment of the present application, the strategy acquisition module 203 is also used to determine whether the initial similarity value is greater than the standard similarity value; if it is greater, a corresponding recommended advertising feature is generated for each advertising media in multiple recommended delivery strategies according to the first media portrait, so as to generate multiple recommended delivery strategies based on the advertising recommendation feature.
[0070] In an optional implementation method of the embodiment of the present application, the media portrait acquisition module 201 acquires media portrait data, which includes media advertising unit price, historical delivery effect, and audience portrait; the media portrait data is processed by a feature conversion method to obtain training set data, and the feature conversion method includes one-hot encoding; the initial machine learning model is trained by the training data set data to obtain a media portrait model.
[0071] In an optional implementation method of the embodiment of the present application, the strategy acquisition module 203 is also used to obtain multiple historical advertising delivery indicators and multiple advertising media portraits; the multiple historical advertising delivery indicators and multiple advertising media portraits are processed through feature engineering to obtain a feature data set, and the feature engineering includes multi-table feature splicing and wide table feature combination; the feature data set is used to train the initial regression model to generate a solution recommendation model.
[0072] The present application also discloses an electronic device. Figure 3 , Figure 3 The electronic device 300 may include: at least one processor 301 , at least one network interface 304 , a user interface 303 , a memory 305 , and at least one communication bus 302 .
[0073] The communication bus 302 is used to realize the connection and communication between these components.
[0074] The user interface 303 may include a display screen (Display) and a camera (Camera). The optional user interface 303 may also include a standard wired interface and a wireless interface.
[0075] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).
[0076] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.
[0077] Among them, the memory 305 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may optionally be at least one storage device located away from the aforementioned processor 301. Refer to Figure 3 , the memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program of an advertisement delivery strategy generation method.
[0078] exist Figure 3In the electronic device 300 shown, the user interface 303 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 301 can be used to call an application program storing an advertising delivery strategy generation method in the memory 305, and when executed by one or more processors 301, the electronic device 300 executes one or more of the methods described in the above embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, 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 to 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 for the present application.
[0079] 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.
[0080] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of 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 service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.
[0081] The units described 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.
[0082] 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.
[0083] If the 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, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.
[0084] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.
[0085] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art that are not disclosed in the present application. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present application are defined by the claims.
Claims
1. A method for generating an advertisement delivery strategy, characterized in that: include: Acquire multiple advertising media portrait data, input the multiple advertising media portrait data into a media portrait model, and obtain multiple advertising media portraits, wherein the multiple advertising media portraits include advertising media types and advertising media audience portraits; Acquire multiple advertising delivery indicators, each of which has a corresponding delivery indicator weight; The multiple advertising delivery indicators and the multiple advertising media portraits are input into a solution recommendation model to obtain multiple recommended delivery strategies.
2. The method according to claim 1, characterized in that After inputting the multiple advertising delivery indicators and the multiple advertising media portraits into a solution recommendation model to obtain multiple recommended delivery strategies, the method further includes: Obtaining first media portraits and target audience portraits corresponding to the multiple recommended delivery strategies; Calculating an initial similarity value between the first media portrait and the target audience portrait by using a preset portrait parameter weight and a preset similarity algorithm; The multiple recommended delivery strategies are sorted according to the initial similarity values to obtain the recommended delivery strategy list.
3. The method according to claim 2, characterized in that Before sorting the plurality of recommended delivery strategies according to the initial similarity values to obtain the recommended delivery strategy list, the method further includes: Obtaining a product delivery record corresponding to each of the plurality of recommended delivery strategies; Calculating, according to the product delivery record and the multiple advertising delivery indicators, a second similarity value between the product delivered in the product delivery record and the product information in the multiple advertising delivery indicators; Determining whether the second similarity value is greater than a preset product similarity value; If it is greater than, then according to the corresponding relationship between the second similarity value and the initial similarity value, the sum of the second similarity value and the initial similarity value is calculated, and the multiple recommended delivery strategies are sorted according to the calculation result to obtain the recommended delivery strategy list.
4. The method according to claim 3, characterized in that After calculating the sum of the second similarity value and the initial similarity value according to the corresponding relationship between the second similarity value and the initial similarity value, and sorting the plurality of recommended delivery strategies according to the calculation result to obtain the recommended delivery strategy list, the method further includes: Determining whether there are multiple online recommended delivery strategies among the multiple recommended delivery strategies; If so, obtaining the audience historical behavior corresponding to the second similarity value being greater than the preset product similarity value in the multiple online recommendation delivery strategies; The multiple recommended delivery strategies are sorted according to the audience historical behaviors and the multiple advertisement delivery indicators to generate a behavior recommendation list.
5. The method according to claim 2, characterized in that: After calculating the initial similarity value between the first media portrait and the target audience portrait by using the preset portrait parameter weights and the preset similarity algorithm, the method further includes: Determining whether the initial similarity value is greater than a standard similarity value; If it is greater, a corresponding recommended advertisement feature is generated for each advertisement media in the plurality of recommended delivery strategies according to the first media portrait, so as to generate the plurality of recommended delivery strategies based on the advertisement recommendation feature.
6. The method according to claim 1, characterized in that Before acquiring a plurality of advertising media portrait data and inputting the plurality of advertising media portrait data into a media portrait model to obtain a plurality of advertising media portraits, the method further includes: Acquire media portrait data, wherein the media portrait data includes media advertising unit price, historical delivery effect, and audience portrait; Processing the media portrait data by a feature conversion method to obtain training set data, wherein the feature conversion method includes one-hot encoding; The initial machine learning model is trained using the training data set data to obtain the media portrait model.
7. The method according to claim 6, characterized in that Before inputting the multiple advertising delivery indicators and the multiple advertising media portraits into a solution recommendation model to obtain multiple recommended delivery strategies, the method further includes: Obtaining multiple historical advertising delivery indicators and the multiple advertising media portraits; Processing the multiple historical advertising delivery indicators and the multiple advertising media portraits through feature engineering to obtain a feature data set, wherein the feature engineering includes multi-table feature concatenation and wide table feature combination; The feature data set is used to train the initial regression model to generate the solution recommendation model.
8. A terminal device, characterized in that: The terminal device comprises: An acquisition module, used to acquire a first data query instruction from a terminal device; an instruction module, configured to send a second data query instruction to the data storage device according to the first data query instruction, wherein the second data query instruction is used to instruct the data storage device to generate a data query result; The reading module is used to read the data query result line by line, so as to forward the data query result line by line to the terminal device until the terminal device obtains the complete data query result.
9. An electronic device, characterized in that: The electronic device (300) comprises a processor (301), a memory (305), a user interface (303) and a network interface (304), wherein the memory (305) is used to store instructions, the user interface (303) and the network interface (304) are used to communicate with other devices, and the processor (301) is used to execute the instructions stored in the memory (305) so that the electronic device (300) executes the method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method steps according to any one of claims 1 to 7 are performed.